ࡱ>  "@ .?bjbjFF 4x,, D@0###*4448l@*( (:b(iiiD *$RUJN#ieiiiN##;i##iFȴ## 0%R4i"طT0t Q0,x$DD(D@!###iiiNN**Dk4**4PART IV Natural resource exploitation chapter 14 The efficient and optimal use of natural resources The Golden Rule is that there are no golden rules. George Bernard Shaw, Maxims for Revolutionists, in Man and Superman Learning objectives Having read this chapter, the reader should be able to understand the ideas of efficient and optimal allocations of environmental resources recognise the relationship between but also the difference between the concepts of efficency and optimality understand how questions relating to efficient and optimal use of environmental resources over time can be analysed using a class of models known as optimal growth models appreciate the ways in which resource use patterns are linked with sustainability Introduction In this chapter, we construct a framework to ana-lyse the use of natural resources over time. This will provide the basis for our investigations of non-renewable resource depletion and the harvesting of renewable resources that follow in Chapters 15 to18. Our objectives in the present chapter are: to develop a simple economic model, built around a production function in which natural resources are inputs into the production process; to identify the conditions that must be satisfied by an economically efficient pattern of natural resource use over time; to establish the characteristics of a socially optimal pattern of resource use over time in the special case of a utilitarian social welfare function. We shall be constructing a stylised model of the economy in order to address questions about the use of resources. Although the economics of our model are straightforward, some mathematics is required to analyse the model. To keep technical difficulties to a minimum, the main body of text avoids the use of mathematical derivations. It shows the logic behind, and the economic interpretations of, import-ant results. Derivations of results are presented separately in appendices at the end of the chapter. It is not vital to read these appendices to follow the arguments in the chapter, but we strongly recommend that you do read them. The derivations are explained thoroughly. Appendix 14.1 is of particular importance as it takes the reader through a key mathematical technique, dynamic optimisation using the maximum principle. You are also urged to read Appendices 14.2 and 14.3 to see how the results discussed in the text are obtained. Part I A simple optimal resource depletion model 14.1 The economy and its production function We begin by specifying the model used in this chapter. The economy produces a single good, Q, which can be either consumed or invested. Consumption increases current well-being, while investment increases the capital stock, permitting greater consumption in the future. Output is generated through a production function using a single composite non-renewable resource input, R. Beginning in this way, with just one type of natural resource, abstracts from any substitution effects that might take place between different kinds of natural resource. In Chapter 15, we shall see how our conclusions alter when more than one type of natural resource enters the production function. In addition to the non-renewable resource, a second input manufactured capital, K enters the production function, which is written as Q = Q(K, R) (14.1) This states that output has some functional relationship to the quantities of the two inputs that are used, but it does not tell us anything about the particular form of this relationship.1 One possible type of production technology is the Cobb Douglas (CD) form, consisting of the class of functions Q = AKR (14.2) where A,  and  > 0. An alternative form, widely used in empirical analysis, is the constant elasticity of substitution (CES) type, which comprises the family of functional forms EMBED Unknown (14.3) EMBED Unknown The CD and CES forms of production function do not exhaust all possibilities. In this chapter, we shall not be making any assumption as to which type of production function best represents the production technology of an economy, but rather work with a general form that might be CD, might be CES, or might be some other. Which functional form is the correct one is an empirical question, and cannot be answered by theoretical argument alone. 14.2 Is the natural resource essential? The characteristics of an optimal resource depletion path through time will be influenced by whether the natural resource is essential. Essentialness of a resource could mean several things. First, a resource might be essential as a waste disposal and reprocessing agent. Given the ubiquitous nature of waste and the magnitude of the damages that waste can cause, resources do appear to be necessary as waste-processing agents. A resource might also be essential for human psychic satisfaction. Humans appear to need solitude and the aesthetic enjoyment derived from observing or being in natural environments. Thirdly, some resource might be ecologically essential in the sense that some or all of a relevant eco-system cannot survive in its absence. In this chapter, we are concerned with a fourth meaning: whether a resource is directly essential for production. Some resources are undoubtedly essential for specific products for example, crude oil is an essential raw material for the production of petrol and paraffin. But here we are conceptualising resources at a high degree of aggregation, dealing with general classes such as non-renewable and renewable resources. A productive input is defined to be essential if output is zero whenever the quantity of that input is zero, irrespective of the amounts of other inputs used. That is, R is essential if Q = Q(K, R = 0) = 0 for any postive value of K. In the case of the CD production function, R and K are both essential, as setting any input to zero in equation 14.2 results in Q = 0. Matters are not so straightforward with the CES function. We state (but without giving a proof) that if  < 0 then no input is essential, and if  > 0 then all inputs are essential. What is the relevance of this to our study of resource use over time? If we wish to answer questions about the long-run properties of economic systems, the essentialness of non-renewable resources will matter. Since, by definition, non-renewable resources exist in finite quantities it is not possible to use constant and positive amounts of them over infinite horizons. However, if a resource is essential, then we know that production can only be undertaken if some positive amount of the input is used. This seems to suggest that production and consumption cannot be sustained indefinitely if a non-renewable resource is a necessary input to production. However, if the rate at which the resource is used were to decline asymptotically to zero, and so never actually become zero in finite time, then production could be sustained indefinitely even if the resource were essential. Whether output could rise, or at least stay constant over time, or whether it would have to decline towards zero will depend upon the extent to which other resources can be substituted for non-renewable resources and upon the behaviour of output as this substitution takes place. 14.3 What is the elasticity of substitution between K and R? The extent of substitution possibilities is likely to have an important bearing on the feasibility of continuing economic growth over the very long run, given the constraints which are imposed by the natural environment. Let us examine substitution between the non-renewable resource and capital. The elasticity of substitution, , between capital and the non-renewable natural resource (from now on just called the resource) is defined as the proportionate change in the ratio of capital to the resource in response to a proportionate change in the ratio of the marginal products of capital and the resource, conditional on total output Q remaining constant (see Chiang, 1984). That is, EMBED Unknown (14.4) where the partial derivative QR = "Q/"R denotes the marginal product of the resource and QK = "Q/"K denotes the marginal product of capital.3 The elasticity of substitution lies between zero and infinity. Substitution possibilities can be represented diagrammatically. Figure 14.1 shows what are known as production function isoquants. For a given production function, an isoquant is the locus of all combinations of inputs which, when used efficiently, yield a constant level of output. The three isoquants shown in Figure 14.1 each correspond to the level of output, EMBED Unknown, but derive from different production functions. The differing substitution possibilities are reflected in the curvatures of the isoquants. [Figure 14.1 near here] In the case where no input substitution is possible (that is,  = 0), inputs must be combined in fixed proportions and the isoquants will be L-shaped. Production functions of this type, admitting no substitution possibilities, are sometimes known as Leontief functions. They are commonly used in input output models of the economy. At the other extreme, if substitution is perfect ( = (), isoquants will be straight lines. In general, a production function will exhibit an elasticity of substitution somewhere between those two extremes (although not all production functions will have a constant  for all input combinations). In these cases, isoquants will often be convex to the origin, exhibiting a greater degree of curvature the lower the elasticity of substitution, . Some evidence on empirically observed values of the elasticity of substitution between different production inputs is presented later in Box 14.1. For a CES production function, we can also relate the elasticity of substitution to the concept of essentialness. It can be shown (see, for example, Chiang, 1984, p. 428) that  = 1/(1 + ). We argued in the previous section that no input is essential where  < 0, and all inputs are essential where  > 0. Given the relationship between  and , it can be seen that no input is essential where  > 1, and all inputs are essential where  < 1. Where  = 1 (that is,  = 0), the CES production function collapses to the CD form, where all inputs are essential. 14.4 Resource substitutability and the consequences of increasing resource scarcity As production continues throughout time, stocks of non-renewable resources must decline. Continuing depletion of the resource stock will lead to the non-renewable resource price rising relative to the price of capital. The results obtained in Chapter 11 imply that as the relative price of the non-renewable resource rises the resource to capital ratio will fall, thereby raising the marginal product of the resource and reducing the marginal product of capital. How-ever, the magnitude of this substitution effect will depend on the size of the elasticity of substitution. Where the elasticity of substitution is high, only small changes in relative input prices will be necessary to induce a large proportionate change in the quantities of inputs used. Resource scarcity will be of little consequence as the economy is able to replace the scarce resource by the reproducible substitute. Put another way, the constraints imposed by the finiteness of the non-renewable resource stock will bite rather weakly in such a case. On the other hand, low substitution possibilities mean that as resource depletion pushes up the relative price of the resource, the magnitude of the induced substitution effect will be small. Resource scarcity will have more serious adverse effects, as the scope for replacement of the scarce resource by the reproducible substitute is more limited. Where the elasticity of substitution is zero, then no scope exists for such replacement. 14.4.1 The feasibility of sustainable development In Chapter 4, we considered what sustainability might mean, how economists have attempted to incorporate a concern with sustainability into their work, and why one might wish to incorporate sustainability into the set of objectives that society pursues. What we did not discuss there was whether sustainable development is actually possible. To address this question, two things are necessary. First, a criterion of sustainability is required; unless we know what sustainability is it is not possible to judge whether it is feasible. Second, we need to describe the material transformation conditions available to society, now and in the future. These conditions the economys production possibilities determine what can be obtained from the endowments of natural and human-made capital over the relevant time horizon. To make some headway in addressing this question a conventional sustainability criterion will be adopted: non-declining per capita consumption maintained over indefinite time (see Chapter 4). Turning attention to the transformation conditions, it is clear that a large number of factors enter the picture. What is happening to the size of the human population? What kinds of resources are available and in what quantities, and what properties do they possess? What will happen to the state of technology in the future? How will ecosystems be affected by the continuing waste loads being placed upon the environment, and how will ecosystem changes feed back upon productive potential? To make progress, economists typically simplify and narrow down the scope of the problem, and represent the transformation possibilities by making an assumption about the form of an economys production function. A series of results have become established for several special cases, deriving mainly from papers by Dasgupta and Heal (1974), Solow (1974a) and Stiglitz (1974). For the CD and CES functions we have the following. Case A: Output is produced under fully competitive conditions through a CD production function with constant returns to scale and two inputs, a non-renewable resource, R, and manufactured capital, K, as in the following special case of equation 14.2: EMBED Unknown Then, in the absence of technical progress and with constant population, it is feasible to have constant consumption across generations if the share of total output going to capital is greater than the share going to the natural resource (that is, if  > ). Case B: Output is produced under fully competitive conditions through a CES production function with constant returns to scale and two inputs, a non-renewable resource, R, and manufactured capital, K, as in equation 14.3: EMBED Unknown Then, in the absence of technical progress and with constant population, it is feasible to have constant consumption across generations if the elasticity of substitution  = 1/(1 + ) is greater than or equal to one. Case C: Output is produced under conditions in which a backstop technology is ermanently available. In this case, the non-renewable natural resource is not essential. Sustainability is feasible, although there may be limits to the size of the constant consumption level that can be obtained. It is relatively easy to gain some intuitive understanding of these results. For the CD case, although the natural resource is always essential in the sense we described above, if  >  then capital is sufficiently substitutable for the natural resource so that output can be maintained by increasing capital as the depletable resource input diminishes. However, it should be noted that there is an upper bound on the amount of output that can be indefinitely sustained in this case; whether that level is high enough to satisfy  survivability (see Chapter 4) is another matter. For the CES case, if  > 1, then the resource is not essential. Output can be produced even in the absence of the natural resource. The fact that the natural resource is finite does not prevent indefinite production (and consumption) of a constant, positive output. Where  = 1, the CES production function collapses to the special case of CD, and so Case A applies. Where a backstop exists (such as a renewable energy source like wind or solar power, or perhaps nuclear-fusion-based power) then it is always possible to switch to that source if the limited natural resource becomes depleted. We explore this process further in the next chapter. These results assumed that the rate of technical progress and the rate of population growth were both zero. Not surprisingly, results change if one or both of these rates is non-zero. The presence of permanent technical progress increases the range of circumstances in which indefinitely long-lived constant per capita consumption is feasible whereas constant population growth has the opposite effect. However, there are circumstances in which constant per capita consumption can be maintained even where population is growing provided the rate of technical progress is sufficiently large and the share of output going to the resource is sufficiently low. Details of this result are given in Stiglitz (1974). Similarly, for a CES production function, sustained consumption is possible even where  < 1 provided that technology growth is sufficiently high relative to population growth. The general conclusion from this analysis is that sustainability requires either a relatively high degree of substitutability between capital and the resource, or a sufficiently large continuing rate of technical progress or the presence of a permanent backstop technology. Whether such conditions will prevail is a moot point. 14.4.2 Sustainability and the Hartwick rule In our discussion of sustainability in Chapter 4, mention was made of the so-called Hartwick savings rule. Interpreting sustainability in terms of non-declining consumption through time, John Hartwick (1977, 1978) sought to identify conditions for sustainability. He identified two sets of conditions which were sufficient to achieve constant (or, more accurately, non-declining) consumption through time: a particular savings rule, known as the Hartwick rule, which states that the rents derived from an efficient extraction programme for the non-renewable resource are invested entirely in reproducible (that is, physical and human) capital; conditions pertaining to the economys production technology. These conditions are essentially those we described in the previous section, which we shall not repeat here. We shall discuss the implications of the Hartwick rule further in Chapter 19. But three comments about it are worth making at this point. First, the Hartwick rule is essentially an ex post description of a sustainable path. Hence if an economy were not already on a sustainable path, then adopting the Hartwick rule is not sufficient for sustainability from that point forwards. This severely reduces the practical usefulness of the rule. (See Asheim, 1986, and Pezzey, 1996, and Appendix 19.1 in the present book.) Second, even were the economy already on a sustainable path, the Hartwick rule requires that the rents be generated from an efficient resource extraction programme in a competitive economy. Third, even if the Hartwick rule is pursued subject to this qualification, the savings rule itself does not guarentee sustainability. Technology conditions may rule out the existence of a feasible path. As we noted in the previous section, feasibility depends very much upon the extent of substitution possibilities open to an economy. Let us now explore this a little further. 14.4.3 How large are resource substitution possibilities? Clearly, the magnitude of the elasticity of substitution between non-renewable resources and other inputs is a matter of considerable importance. But how large it is cannot be deduced by a priori theoretical reasoning this magnitude has to be inferred empirically. Whereas many economists believe that evidence points to reasonably high substitution possibilities (although there is by no means a consensus on this), natural scientists and ecologists stress the limited substitution possibilities between resources and reproducible capital. Indeed some ecologists have argued that, in the long term, these substitution possibilities are zero. These disagreements reflect, in large part, differences in conceptions about the scope of services that natural resources provide. For example, whereas it appears to be quite easy to economise on the use of fossil energy inputs in many production processes, reproducible capital cannot substitute for natural capital in the provision of the amenities offered by wilderness areas, or in the regulation of the earths climate. The reprocessing of harmful wastes is less clear-cut; certainly reproducible capital and labour can substitute for the waste disposal functions of the environment to some extent (perhaps through increased use of recycling processes) but there appear to be limits to how far this substitution can proceed. Finally, it is clear that even if we were to establish that substitutability had been high in the past, this does not imply that it will continue to be so in the future. It may be that as development pushes the economy to points where natural constraints begin to bite, substitution possibilities reduce significantly. Recent literature from natural science seems to suggest this possibility. On the other hand, a more optimistic view is suggested by the effect of technological progress, which appears in many cases to have contributed towards enhanced opportunities for substitution. You should now read the material on resource substitutability presented in Box 14.1. Box 14.1 Resource substitutability: one item of evidence A huge amount of empirical research has been devoted to attempts to measure the elasticity of substitution between particular pairs of inputs. Results of these exercises are often difficult to apply to general models of the type we use in this chapter, because the estimates tend to be specific to the particular contexts being studied, and because many studies work at a much more disaggregated level than is done here. We restrict comments to just one estimate, which has been used in a much-respected model of energyenvironment interactions in the United States economy. Alan Manne, in developing the ETA Macro model, considers a production function in which gross output (Q) depends upon four inputs: K, L, E and N (respectively capital, labour, electric and non-electric energy). Mannes production function incorporates the following assumptions: (a) There are constant returns to scale in terms of all four inputs. (b) There is a unit elasticity of substitution between capital and labour. (c) There is a unit elasticity of substitution between electric and non-electric energy. (d) There is a constant elasticity of substitution between the two pairs of inputs, capital and labour on the one hand and electric and non-electric energy on the other. Denoting this constant elasticity of substitution by the symbol , the production function used in the ETA Macro model that embraces these assumptions is EMBED Unknown where, as noted in the text, EMBED Unknown Manne selects the value 0.25, a relatively low figure, for the elasticity of substitution s between the pair of energy inputs and the other input pair. How is this figure arrived at? First, Manne argues that the elasticity of substitution is approximately equal to the absolute value of the price elasticity of demand for primary energy (see Hogan and Manne, 1979). Then, Manne collects time-series data on the prices of primary energy, incomes and quantities of primary energy consumed. This permits a statistically derived estimate of the long-run price elasticity of demand for primary energy to be obtained, thereby giving an approximation to the elasticity of substitution between energy and other production inputs. Mannes elasticity estimate of 0.25 falls near the median of recent econometric estimates of this elasticity of substitution. Being positive, this figure suggests that energy demand will rise relative to other input demand if the relative price of other inputs to energy rises, and so the composite energy resource is a substitute for other productive inputs (a negative sign would imply the pair were complements). However, as the absolute value of the elasticity is much less than one, the degree of substitutability is very low, implying that relative input demands will not change greatly as relative input prices change. Source: Manne (1979) Up to this point in our presentation, natural re-sources have been treated in a very special way. We have assumed that there is a single, non-renewable resource, R, of fixed, known size, and (implicitly) of uniform quality. Substitution possibilities have been limited to those between this resource and other, non-natural, resources. In practice, there are a large number of different natural resources, with substitution possibilities between members of this set. Of equal importance is the non-uniform quality of resource stocks. Resource stocks do not usually exist in a fixed amount of uniform quality, but rather in deposits of varying grade and quality. As high-grade reserves become exhausted, extraction will turn to lower-grade deposits, provided the resource price is sufficiently high to cover the higher extraction costs of the lower-grade mineral. Furthermore, while there will be some upper limit to the physical occurrence of the resource in the earths crust, the location and extent of these deposits will not be known with certainty. As known reserves become depleted, exploration can, therefore, increase the size of available reserves. Finally, renewable resources can act as backstops for non-renewable: wind and wave power are substitutes for fossil fuels, and wood products are substitutes for metals for some construction purposes, for example. Dasgupta (1993) examines these various substitution possibilities. He argues that they can be classified into nine innovative mechanisms: an innovation allowing a given resource to be used for a given purpose. An example is the use of coal in refining pig-iron; the development of new materials, such as synthetic fibres; technological developments which increase the productivity of extraction processes. For example, the use of large-scale earthmoving equipment facilitating economically viable strip-mining of low-grade mineral deposits; scientific and technical discovery which makes exploration activities cheaper. Examples include developments in aerial photography and seismology; technological developments that increase efficiency in the use of resources. Dasgupta illustrates this for the case of electricity generation: between 1900 and the 1970s, the weight of coal required to produce one kilowatt-hour of electricity fell from 7 lb to less than 1 lb; development of techniques which enable one to exploit low-grade but abundantly available deposits. For example, the use of froth-flotation, allowing low-grade sulphide ores to be concentrated in an economical manner; constant developments in recycling techniques which lower costs and so raise effective resource stocks; substitution of low-grade resource reserves for vanishing high-grade deposits; substitution of fixed manufacturing capital for vanishing resources. In his assessment of substitution possibilities, Dasgupta (p. 1126) argues that only one of these nine mechanisms is of limited scope, the substitution of fixed manufacturing capital for natural resources: Such possibilities are limited. Beyond a point fixed capital in production is complementary to resources, most especially energy resources. Asymptotically, the elasticity of substitution is less than one. There is a constant tension between forces which raise extraction and refining costs the depletion of high-grade deposits and those which lower such costs discoveries of newer technological processes and materials. What implications does this carry for resource scarcity? Dasgupta argues that as the existing resource base is depleted, profit opportunities arise from expanding that resource base; the expansion is achieved by one or more of the nine mech-anisms just described. Finally, in a survey of the current stocks of mineral resources, Dasgupta notes that after taking account of these substitution mechanisms, and assuming unchanged resource stock to demand ratios: the only cause for worry are the phosphates (a mere 1300 years of supply), fossil fuels (some 2500 years), and manganese (about 130000 years). The rest are available for more than a million years, which is pretty much like being inexhanstible. However, adjusting for population and income growth, the supply of hydrocarbons. . .will only last a few hundred years. . .So then, this is the fly in the ointment, the bottleneck, the binding constraint. Dasguptas optimism is not yet finished. He conjectures that profit potentials will induce technological advances (perhaps based on nuclear energy, perhaps on renewables) that will overcome this binding constraint. Not all commentators share this sanguine view, as we have seen previously, and we shall have more to say about resource scarcity in the next chapter. In the meantime, we return to our simple model of the economy, in which the heterogeneity of resources is abstracted from, and in which we conceive of there being one single, uniform, natural resource stock. 14.5 The social welfare function and an optimal allocation of natural resources Chapters 5 and 11 established the meaning of the concepts of efficiency and optimality for the allocation of productive resources in general. We shall now apply these concepts to the particular case of natural resources. Our objective is to establish what conditions must be satisfied for natural resource allocation to be optimal, in the sense that the allocation maximises a social welfare function. The presentation in this chapter focuses upon non-renewable resources, although we also indicate how the ideas can be applied to renewable resources. The first thing we require is a social welfare function. You already know that a general way of writing the social welfare function (SWF) is: W = W (U0, U1, U2,. . ., UT) (14.5) where Ut, t = 0,. . ., T, is the aggregate utility in period t.4 We now assume that the SWF is utilitarian in form. A utilitarian SWF defines social welfare as a weighted sum of the utilities of the relevant individuals. As we are concerned here with inter-temporal welfare, we can interpret an individual to mean an aggregate of persons living at a certain point in time, and so refer to the utility in period 0, in period 1, and so on. Then a utilitarian inter-temporal SWF will be of the form W = 0U0 ++  1U1+  2U2 +... +  TUT (14.6) Now let us assume that utility in each period is a concave function of the level of consumption in that period, so that Ut = U(Ct) for all t, with UC > 0 and UCC < 0. Notice that the utility function itself is not dependent upon time, so that the relationship between consumption and utility is the same in all periods. By interpreting the weights in equation 14.6 as discount factors, related to a social utility discount rate  that we take to be fixed over time, the social welfare function can be rewritten as EMBED Unknown (14.7) For reasons of mathematical convenience, we switch from discrete-time to continuous-time notation, and assume that the relevant time horizon is infinite. This leads to the following special case of the utilitarian SWF: EMBED Unknown (14.8) There are two constraints that must be satisfied by any optimal solution. First, all of the resource stock is to be extracted and used by the end of the time horizon (as, after this, any remaining stock has no effect on social well-being). Given this, together with the fact that we are considering a non-renewable resource for which there is a fixed and finite initial stock, the total use of the resource over time is constrained to be equal to the fixed initial stock. Denoting the initial stock (at t = 0) as S0 and the rate of extraction and use of the resource at time t as Rt, we can write this constraint as EMBED Unknown (14.9) Notice that in equation 14.9, as we are integrating over a time interval from period 0 to any later point in time t, it is necessary to use another symbol (here , the Greek letter tau) to denote any point in time in the range over which the function is being integrated. Equation 14.9 states that the stock remaining at time t (St) is equal to the magnitude of the initial stock (S0) less the amount of the resource extracted over the time interval from zero to t (given by the integral term on the right-hand side of the equation). An equivalent way of writing this resource stock constraint is obtained by differentiating equation 14.9 with respect to time, giving EMBED Unknown (14.10) where the dot over a variable indicates a time derivative, so that EMBED Unknown. Equation 14.10 has a straightforward interpretation: the rate of depletion of the stock, EMBED Unknown, is equal to the rate of resource stock extraction, Rt. A second constraint on welfare optimisation derives from the accounting identity relating consumption, output and the change in the economys stock of capital. Output is shared between consumption goods and capital goods, and so that part of the economys output which is not consumed results in a capital stock change. Writing this identity in continuous-time form we have5 EMBED Unknown (14.11) It is now necessary to specify how output, Q, is determined. Output is produced through a production function involving two inputs, capital and a non-renewable resource: Qt = Q(Kt, Rt) (14.12) Substituting for Qt in equation 14.11 from the production function 14.12, the accounting identity can be written as EMBED Unknown (14.13) We are now ready to find the solution for the socially optimal intertemporal allocation of the non-renewable resource. To do so, we need to solve a constrained optimisation problem. The objective is to maximise the economys social welfare function subject to the non-renewable resource stockflow constraint and the national income identity. Writing this mathematically, therefore, we have the following problem: Select values for the choice variables Ct and Rt for t = 0,. . ., ( so as to maximize EMBED Unknown subject to the constraints EMBED Unknown and EMBED Unknown The full solution to this constrained optimisation problem, and its derivation, are presented in Appendix 14.2. This solution is obtained using the maximum principle of optimal control. That technique is explained in Appendix 14.1, which you are recommended to read now. Having done that, then read Appendix 14.2, where we show how the maximum principle is used to solve the problem that has been posed in the text. If you find this appendix material difficult, note that the text of this chapter has been written so that it can be followed without having read the appendices. In the following sections, we outline the nature of the solution, and provide economic interpretations of the results. 14.5.1 The nature of the solution Four equations characterise the optimal solution: EMBED Unknown (14.14a) EMBED Unknown (14.14b) EMBED Unknown (14.14c) EMBED Unknown (14.14d) Before we discuss the economic interpretations of these equations, it is necessary to explain several things about the notation used and the nature of the solution: The terms QK (= "Q/"K) and QR (= "Q/"R) are the partial derivatives of output with respect to capital and the non-renewable resource. In economic terms, they are the marginal products of capital and the resource, respectively. Time subscripts are attached to these marginal products to make explicit the fact that their values will vary over time in the optimal solution. The terms Pt and t are the shadow prices of the two productive inputs, the natural resource and capital. These two variables carry time subscripts because the shadow prices will vary over time. The solution values of Pt and t, for t = 0, 1, . . ., (, define optimal time paths for the prices of the natural resource and capital.6 The quantity being maximised in equation 14.8 is a sum of (discounted) units of utility. Hence the shadow prices are measured in utility, not consumption (or money income), units. You should now turn to Box 14.2 where an explanation of the relationship between prices in utils and prices in consumption (or income) units is given. Box 14.2 Prices in units of utility: what does this mean? The notion of prices being measured in units of utility appears at first sight a little strange. After all, we are used to thinking of prices in units of money: a Cadillac costs $40000, a Mars bar 30 pence, and so on. Money is a claim over goods and services: the more money someone has, the more goods he or she can consume. So it is evident that we could just as well describe prices in terms of consumption units as in terms of money units. For example, if the price of a pair of Levi 501 jeans were $40, and we agree to use that brand of jeans as our standard commodity, then a Cadillac will have a consumption units price of 1000. We could be even more direct about this. Money can itself be thought of as a good and, by convention, one unit of this money good has a price of one unit. The money good serves as a numeraire in terms of which the relative prices of all other goods are expressed. So one pair of Levis has a consumption units price of 40, or a money price of $40. What is the conclusion of all this? Essentially, it is that prices can be thought of equally well in terms of consumption units or money units. They are alternative but equivalent ways of measuring some quantity. Throughout this book, the terms benefits and costs are usually measured in units of money or its consumption equivalent. But we sometimes refer to prices in utils units of utility rather than in money/consumption units. It is here that matters may be a little baffling. But this turns out to be a very simple notion. Economists make extensive use of the utility function: U = U(C) where U is units of utility and C is units of consumption. Now suppose that the utility function were of the simple linear form U = kC where k is some constant, positive number. Then units of utility are simply a multiple of units of consumption. So if k = 2, three units of consumption are equivalent to six units of utility, and so on. But the utility function may be non-linear. Indeed, it is often assumed that utility rises with consumption but at a decreasing rate. One form of utility function that satisfies this assumption is U = log (C + 1), with log denoting the common logarithmic operator, and in which the argument of the function includes the additive constant 1 to keep utility non-negative. The chart in Figure 14.2 shows the relationship between utility and consumption for this particular utility function. [Figure 14.2 near here] It is equally valid to refer to prices in utility units as in any other units. From Figure 14.2 it is clear that a utility price of 2 corresponds to a consumption (or money) price of approximately 100 (in fact, 99 exactly). Also, a consumption price of 999 corresponds to a utility price of 3. What is the consumption units price equivalent to a price of 2.5 units of utility? (Use a calculator to find the answer, or read it off approximately from the diagram.) Which units prices are measured in will depend on how the problem has been set up. In the chapters on resource depletion, what is being maximised is social welfare; given that the SWF is specified as a sum of utilities (of different people or different generations), it seems natural to denominate it in utility units as well, although our discussion makes it clear that we could convert units from utility into money/consumption terms if we wished to do so. In other parts of the book, what is being maximised is net benefit. That measure is typically constructed in consumption (or money income) units, and so it is natural to use money prices when dealing with problems set up in this way. In conclusion, it is up to us to choose which units are most convenient. And provided we know what the utility function is (or are willing to make an assumption about what it is) then we can always move from one to the other as the occasion demands. Now we are in a position to interpret the four conditions 14.14. First recall from the discussions in Chapters 5 and 11 that for any resource to be allocated efficiently, two kinds of conditions must be satisfied: static and dynamic efficiency. The first two of these conditions 14.14a and 14.14b are the static efficiency conditions that arise in this problem; the latter two are the dynamic efficiency conditions which must be satisfied. These are examined in a moment. The first two conditions 14.14a and 14.14b also implicitly define an optimal solution to this problem. We shall explain what this means shortly. 14.5.2 The static and dynamic efficiency conditions You will recall from our discussions in Chapters 5 and 11 that the efficient allocation of any resource consists of two aspects. 14.5.2.1 The static efficiency conditions As with any asset, static efficiency requires that, in each use to which a resource is put, the marginal value of the services from it should be equal to the marginal value of that resource stock in situ. This ensures that the marginal net benefit (or marginal value) to society of the resource should be the same in all its possible uses. Inspection of equations 14.14a and 14.14b shows that this is what these equations do imply. Look first at equation 14.14a. This states that, in each period, the marginal utility of consumption UC,t must be equal to the shadow price of capital t (remembering that prices are measured in units of utility here). A marginal unit of output can be used for consumption now (yielding UC,t units of utility) or added to the capital stock (yielding an amount of capital value t in utility units). An efficient outcome will be one in which the marginal net benefit of using one unit of output for consumption is equal to its marginal net benefit when it is added to the capital stock. Equation 14.14b states that the value of the marginal product of the natural resource must be equal to the marginal value (or shadow price) of the natural resource stock. This shadow price is, of course, Pt. The value of the marginal product of the resource is the marginal product in units of output (i.e. QR,t) multiplied by the value of one unit of output, t. But we have defined t as the price of a unit of capital; so why is this the value of one unit of output? The reason is simple. In this economy, units of output and units of capital are in effect identical (along an optimal path). Any output that is not consumed is added to capital. So we can call t either the value of a marginal unit of capital or the value of a marginal unit of output. 14.5.2.2 The dynamic efficiency conditions Dynamic efficiency requires that each asset or resource earns the same rate of return, and that this rate of return is the same at all points in time, being equal to the social rate of discount. Equations 14.14c and 14.14d ensure that dynamic efficiency is satisfied. Consider first equation 14.14c. Dividing each side by P we obtain EMBED Unknownwhich states that the growth rate of the shadow price of the natural resource (that is, its own rate of return) should equal the social utility discount rate. Finally, dividing both sides of 14.14d by , we obtain EMBED Unknown which states that the return to physical capital (its capital appreciation plus its marginal productivity) must equal the social discount rate. 14.5.3 Hotelling s rule: two interpretations Equation 14.14c is known as Hotellings rule for the extraction of non-renewable resources. It is often expressed in the form EMBED Unknown (14.15) The Hotelling rule is an intertemporal efficiency condition which must be satisfied by any efficient process of resource extraction. In one form or another, we shall return to the Hotelling rule during this and the following three chapters. One interpretation of this condition was offered above. A second can be given in the following way. First rewrite equation 14.15 in the form used earlier, that is EMBED Unknown (14.16) By integration of equation 14.16 we obtain Pt = P0e(t (14.17) Pt is the undiscounted price of the natural resource. The discounted price is obtained by discounting Pt at the social utility discount rate (. Denoting the discounted resource price by P*, we have EMBED Unknown (14.18) Equation 14.18 states that the discounted price of the natural resource is constant along an efficient resource extraction path. In other words, Hotellings rule states that the discounted value of the resource should be the same at all dates. But this is merely a special case of a general assetefficiency condition; the discounted (or present value) price of any efficiently managed asset will remain constant over time. This way of interpreting Hotellings rule shows that there is nothing special about natural resources per se when it comes to thinking about efficiency. A natural resource is an asset. All efficiently managed assets will satisfy the condition that their discounted prices should be equal at all points in time. If we had wished to do so, the Hotelling rule could have been obtained directly from this general condition. Before moving on, note the effect of changes in the social discount rate on the optimal path of resource price. The higher is , the faster should be the rate of growth of the natural resource price. This result is eminently reasonable given the two interpretations we have offered of the Hotelling rule. Its implications will be explored in the following chapter. 14.5.4 The growth rate of consumption We show in Appendix 14.2 that equations 14.14a and 14.14d can be combined to give EMBED Unknown The term , the elasticity of marginal utility with respect to consumption, is necessarily positive under the assumptions we have made about the utility function. It therefore follows that EMBED Unknown Some intuition may help to understand these relations. The social discount rate, , reflects impatience for future consumtion; QK (the marginal product of capital) is the pay-off to delayed consumption. The relations imply that along an optimal path: (a) consumption is increasing when pay-off is greater than impatience; (b) consumption is constant when pay-off is equal to impatience; (c) consumption is decreasing when pay-off is less than impatience. Therefore, consumption is growing over time along an optimal path if the marginal product of capital (QK) exceeds the social discount rate (); consumption is constant if QK = ; and consumption growth is negative if the marginal product of capital is less than the social discount rate. This makes sense, given that: (a) when  pay-off is greater than  impatience, the economy will be accumulating K and hence growing; (b) when pay-off and impatience are equal, K will be constant; (c) when pay-off is less than impatience, the economy will be running down K. 14.5.5 Optimality in resource extraction The astute reader will have noticed that we have described the Hotelling rule (and the other conditions described above) as an efficiency condition. But a rule that requires the growth rate of the price of a resource to be equal to the social discount rate does not give rise to a unique price path. This should be evident by inspection of Figure 14.3, in which two different initial prices, say 1 util and 2 utils, grow over time at the same discount rate, say 5%. If  were equal to 5%, each of these paths  and indeed an infinite quantity of other such price paths  satisfies Hotelling s rule, and so they are all efficient paths. But only one of these price paths can be optimal, and so the Hotelling rule is a necessary but not a sufficient condition for optimality. [Figure 14.3 near here] How do we find out which of all the possible efficient price paths is the optimal one? An optimal solution requires that all of the conditions listed in equations 14.14ad, together with initial conditions relating to the stocks of capital and resources and terminal conditions (how much stocks, if any, should be left at the terminal time), are satisfied simultaneously. So Hotellings rule one of these conditions is a necessary but not sufficient condition for an optimal allocation of natural resources over time. Let us think a little more about the initial and final conditions for the natural resource that must be satisfied. There will be some initial resource stock; similarly, we know that the resource stock must converge to zero as elapsed time passes and the economy approaches the end of its planning horizon. If the initial price were too low, then this would lead to too large amounts of resource use in each period, and all the resource stock would become depleted in finite time (that is, before the end of the planning horizon). Conversely, if the initial price were too high, then this would lead to too small amounts of resource use in each period, and some of the resource stock would (wastefully) remain undepleted at the end of the planning horizon. This suggests that there is one optimal initial price that would bring about a path of demands that is consistent with the resource stock just becoming fully depleted at the end of the planning period. In conclusion, we can say that while equations 14.14 are each efficiency conditions, taken jointly as a set (together with initial values for K and S) they implicitly define an optimal solution to the optimisation problem, by yielding unique time paths for Kt and Rt and their associated prices that maximise the social welfare function. Part II Extending the model to incorporate extraction costs and renewable resources In our analysis of the depletion of resources up to this point, we have ignored extraction costs. Usually the extraction of a natural resource will be costly. So it is desirable to generalise our analysis to allow for the presence of these costs. It seems likely that total extraction costs will rise as more of the material is extracted. So, denoting total extraction costs as ( and the amount of the resource being extracted as R, we would expect that ( will be a function of R. A second influence may also affect extraction costs. In many circumstances, costs will depend on the size of the remaining stock of the resource, typically rising as the stock becomes more depleted. Letting St denote the size of the re-source stock at time t (the amount remaining after all previous extraction) we can write extraction costs as EMBED Unknown (14.19) To help understand what the presence of the stock term in equation 14.19 implies about extraction costs look at Figure 14.4. This shows three possible relationships between total extraction costs and the remaining resource stock size for a constant level of resource extraction. The relationship denoted (i) corresponds to the case where the total extraction cost is independent of the stock size. In this case, the extraction cost function collapses to the simpler form (t = (1(Rt) in which extraction costs depend only on the quantity extracted per period of time. In case (ii), the costs of extracting a given quantity of the resource increase linearly as the stock becomes increasingly depleted. (S = "(/"S is then a constant negative number. Finally, case (iii) shows the costs of extracting a given quantity of the resource increasing at an increasing rate as S falls towards zero; (S is negative but not constant, becoming larger in absolute value as the resource stock size falls. This third case is the most likely one for typical non-renewable resources. Consider, for example, the cost of extracting oil. As the available stock more closely approaches zero, capital equipment is directed to exploiting smaller fields, often located in geographically difficult land or marine areas. The quality of resource stocks may also fall in this process, with the best fields having been exploited first. These and other similar reasons imply that the cost of extracting an additional barrel of oil will tend to rise as the remaining stock gets closer to exhaustion. 14.6 The optimal solution to the resource depletion model incorporating extraction costs The problem we now wish to solve can be expressed as follows: Select values for the choice variables Ct and Rt for t = 0,. . ., (( so as to maximise EMBED Unknown subject to the constraints EMBED Unknown and EMBED Unknown Comparing this with the description of the optimisation problem for the simple model, one difference can be found. The differential equation for D now includes extraction costs as an additional term. Output produced through the production function Q(K, R) should now be thought of as gross output. Extraction costs make a claim on this gross output, and net-of-extraction cost output is gross output minus extraction costs (that is, Q ((). The solution to this problem is once again obtained using the maximum principle of optimal control. If you wish to go through the derivations, you should follow the steps in Appendix 14.2, but this time ensuring that you take account of the extraction cost term which will now appear in the differential equation for D, and so also in the Hamiltonian. From this point onwards, we shall omit time subscripts for simplicity of notation unless their use is necessary in a particular context. The necessary conditions for a social welfare optimum now become: EMBED Unknown (14.20a) EMBED Unknown (14.20b) EMBED Unknown (14.20c) EMBED Unknown (14.20d) Note that two of these four equations 14.20a and 14.20d are identical to their counterparts in the solution to the simple model we obtained earlier, and the interpretations offered then need not be repeated. However, the equations for the resource net price and for the rate of change of net price differ. Some additional comment is warranted on these two equations. First, it is necessary to distinguish between two kinds of price: gross price and net price. This distinction follows from that we have just made between gross and net output. These two measures of the resource price are related by net price being equal to gross price less marginal cost. Equation 14.20b can be seen in this light: Pt = tQR less tR Net price = Gross price less Marginal cost The term tR is the value of the marginal extraction cost, being the product of the impact on output of a marginal change in resource extraction and the price of capital (which, as we saw earlier, is also the value of one unit of output). Equation 14.20b can be interpreted in a similar way to that given for equation 14.14b. That is, the value of the marginal net product of the natural resource ((QR ((R, the marginal gross product less the marginal extraction cost) must be equal to the marginal value (or shadow net price) of a unit of the natural resource stock, P. If profit-maximising firms in a competitive economy were extracting resources, these marginal costs would be internal to the firm and the market price would be identical to the gross price. Note that the level of the net (and the gross) price is only affected by the effect of the extraction rate, R, on costs. The stock effect does not enter equation 14.20b. The stock effect on costs does, however, enter equation 14.20c for the rate of change of the net price of the resource. This expression is the Hotelling rule, but now generalised to take account of extraction costs. The modified Hotelling rule (equation 14.20c) is: EMBED Unknown Given that S = "/"S is negative (resource extraction is more costly the smaller is the remaining stock), efficient extraction over time implies that the rate of increase of the resource net price should be lower where extraction costs depend upon the resource stock size. A little reflection shows that this is eminently reasonable. Once again, we work with an interpretation given earlier. Dividing equation 14.20c by the resource net price we obtain which says that, along an efficient price path, the social rate of discount should equal the rate of return from holding the resource (which is given by its rate of capital appreciation, plus the present value of the extraction cost increase that is avoided by not extracting an additional unit of the stock,  S/P.) There is yet another possible interpretation of equation 14.20c. To obtain this, first rearrange the equation to the form: EMBED Unknown (14.20*) The left-hand side of 14.20* is the marginal cost of not extracting an additional unit of the resource; the right-hand side is the marginal benefit from not extracting an additional unit of the resource. At an efficient (and at an optimal) rate of resource use, the marginal costs and benefits of resource use are balanced at each point in time. How is this interpretation obtained? Look first at the left-hand side. The net price of the resource, P, is the value that would be obtained by the resource owner were he or she to extract and sell the resource in the current period. With  being the social utility discount rate, P is the utility return forgone by not currently extracting one unit of the resource, but deferring that extraction for one period. This is sometimes known as the holding cost of the resource stock. The right-hand side contains two components. EMBED Unknownis the capital appreciation of one unit of the unextracted resource; the second component, S, is a return in the form of a postponement of a cost increase that would have occurred if the additional unit of the resource had been extracted. Finally, note that whereas the static efficiency condition 14.20b is only affected by the current extraction rate, R, the dynamic efficiency condition (Hotellings rule, 14.20c) is only affected by the stock effect on costs. In conclusion, the presence of costs related to the level of resource extraction raises the gross price of the resource above its net price but has no effect on the growth rate of the resource net price. Note that net price is what we referred to as rent in Chapter 4: it is also sometimes referred to as royalty. In contrast, a resource stock size effect on extraction costs will slow down the rate of growth of the resource net price. In most circumstances, this implies that the resource net price has to be higher initially (but lower ultimately) than it would have been in the absence of this stock effect. As a result of higher initial prices, the rate of extraction will be slowed down in the early part of the time horizon, and a greater quantity of the resource stock will be conserved (to be extracted later). 14.7 Generalisation to renewable resources We reserve a lengthy analysis of the allocation of renewable resources until Chapters 17 and 18, but it will be useful at this point to suggest the way in which the analysis can be undertaken. To do so, first note that we have been using S to represent a fixed and finite stock of a non-renewable resource. The total use of the resource over time was constrained to be equal to the fixed initial stock. This relationship arises because the natural growth of non-renewable resources is zero except over geological periods of time. Thus we wrote EMBED Unknown However, the natural growth of renewable resources is, in general, non-zero. A simple way of modelling this growth is to assume that the amount of growth of the resource, Gt, is some function of the current stock level, so that Gt = G(St). Given this we can write the relationship between the change in the resource stock and the rate of extraction (or harvesting) of the resource as EMBED Unknown (14.21) Not surprisingly, the efficiency conditions required by an optimal allocation of resources are now dif-ferent from the case of non-renewable resources. However, a modified version of the Hotelling rule for rate of change of the net price of the resource still applies, given by EMBED Unknown (14.22) where GS = dG/dS, and in which we have assumed, for simplicity, that harvesting does not incur costs, nor that any natural damage results from the harvesting and use of the resource. Inspection of the modified Hotelling rule for renewable resources (equation 14.22) demonstrates that the rate at which the net price should change over time depends upon GS, the rate of change of resource growth with respect to changes in the stock of the resource. We will not attempt to interpret equation 14.22 here as that is best left until we examine renewable resources in detail later. However, it is worth saying a few words about steady-state resource harvesting. A steady-state harvesting of a renewable resource exists when all stocks and flows are constant over time. In particular, a steady-state harvest will be one in which the harvest level is fixed over time and is equal to the natural amount of growth of the resource stock. Additions to and subtractions from the resource stock are thus equal, and the stock remains at a constant level over time. Now if the demand for the resource is constant over time, the resource net price will remain constant in a steady state, as the quantity harvested each period is constant. Therefore, in a steady state, EMBED UnknownSo in a steady state, the Hotelling rule simplifies to P = PGS (14.23) and so  = GS (14.24) It is common to assume that the relationship between the resource stock size, S, and the growth of the resource, G, is as indicated in Figure 14.5. This relationship is explained more fully in Chapter 17. As the stock size increases from zero, the amount of growth of the resource rises, reaches a maximum, known as the maximum sustainable yield (MSY), and then falls. Note that GS = dG/dS is the slope at any point of the growthstock function in Figure 14.5. [Figure 14.5 near here] We can deduce that if the social utility discount rate  were equal to zero, then the efficiency condition of equation 14.24 could only be satisfied if the steady-state stock level is \, and the harvest is the MSY harvest level. On the other hand, if the social discount rate was positive (as will usually be the case), then the efficiency condition requires that the steady-state stock level be less than \. At the stock level S*, for example, GS is positive, and would be an efficient stock level, yielding a sustainable yield of G*, if the discount rate were equal to this value of GS. Full details of the derivation of this and other results relating to the Hotelling rule are given in Appendix 14.3. 14.8 Complications The model with which we began in this chapter was one in which there was a single, known, finite stock of a non-renewable resource. Furthermore, the whole stock was assumed to have been homogeneous in quality. In practice, both of these assumptions are often false. Rather than there existing a single non-renewable resource, there are many different classes or varieties of non-renewable resource, some of which may be relatively close substitutes for others (such as oil and natural gas, and iron and aluminium). While it may be correct to assume that there exists a given finite stock of each of these resource stocks in a physical sense, the following situations are likely: The total stock is not known with certainty. New discoveries increase the known stock of the resource. A distinction needs to be drawn between the physical quantity of the stock and the economically viable stock size. Research and development, and technical progress, take place, which can change extraction costs, the size of the known resource stock, the magnitude of economically viable resource deposits, and estimates of the damages arising from natural resource use. Furthermore, even when we focus on a particular kind of non-renewable resource, the stock is likely to be heterogeneous. Different parts of the total stock are likely to be uneven in quality, or to be located in such a way that extraction costs differ for different portions of the stock. By treating all non-renewable resources as one composite good, our analysis in this chapter had no need to consider substitutes for the resource in question (except, of course, substitutes in the form of capital and labour). But once our analysis enters the more complex world in which there are a variety of different non-renewable resources which are substitutable for one another to some degree, analysis inevitably becomes more complicated. One particular issue of great potential importance is the presence of backstop technologies (see Chapter 15). Suppose that we are currently using some non-renewable resource for a particular purpose perhaps for energy production. It may well be the case that another resource exists that can substitute entirely for the resource we are considering, but may not be used at present because its cost is relatively high. Such a resource is known as a backstop technology. For example, renewable power sources such as wind energy are backstop alternatives to fossil-fuel-based energy. The existence of a backstop technology will set an upper limit on the level to which the price of a resource can go. If the cost of the original resource were to exceed the backstop cost, users would switch to the backstop. So even though renewable power is probably not currently economically viable, at least not on a large scale, it would become so at some fossil-fuel cost, and so the existence of a backstop will lead to a price ceiling for the original resource. Each of the issues we have raised in this section, and which we have collectively called complications, need to be taken account of in any comprehensive account of resource use. We shall do so in the next four chapters. 14.9 A numerical application: oil extraction and global optimal consumption In this section we present a simple, hypothetical numerical application of the theory developed above. You may find the mathematics of the solution given in Box 14.3 a little tedious; if you wish to avoid the maths, just skip the box and proceed to Table 14.1 and Figures 14.614.8 at the end of the section where the results are laid out. (The derivation actually uses the technique of dynamic optimisation explained in Appendix 14.1, but applied in this case to a discrete-time model.) [Figure 14.6 near here] [Figure 14.7 near here] [Figure 14.8 near here] Box 14.3 Solution of the dynamic optimisation problem using the maximum principle The current value of the Hamiltonian is Ht = U(Ct) + Pt+1(Rt) + qt+1(Q(Kt, Rt) Ct) (t = 0, 1,. . ., T 1) where Pt is the shadow price of oil (at time t), and qt is the shadow price of capital. The four necessary conditions for an optimum are: EMBED Unknown which implies Hotellings efficiency rule EMBED Unknown EMBED Unknown which implies EMBED Unknown where EMBED Unknown EMBED Unknown where EMBED Unknown EMBED Unknown where EMBED Unknown Since we know ST = KT = 0, there are 6T unknowns in this problem as given below. The unknowns are: Number T + 1 T + 1 T 1 T 1 Unknowns EMBED Unknown EMBED Unknown We have 6T equations to solve for these 6T unknowns. The equations are: EMBED Unknown [T equations] Pt+1 = qt+1QRt (t = 0, 1,. . ., T 1) [T equations] St+1 = St Rt (t = 0, 1,. . ., T 1) [T equations] Pt+1 = (1 + )Pt (t = 1,. . ., T) [T equations] EMBED Unknown [T equations] Kt+1 = Kt + Q(Kt, Rt)  Ct (t = 0, 1,. . ., T  1) [T equations] Suppose that the welfare function to be maximised is EMBED Unknown where Ct is the global consumption of goods and services at time t; U is the utility function, with U(Ct) = log Ct;  is the utility discount rate; and T is the terminal point of the optimisation period. The relevant constraints are EMBED Unknown S0 and K0 are given. St denotes the stock of oil; Rt is the rate of oil extraction; Kt is the capital stock; and Q(Kt, Rt) = AKt0.9Rt0.1 is a CobbDouglas production function, with A being a fixed efficiency parameter. In this application, we assume that oil extraction costs are zero, and that there is no depreciation of the capital stock. Note that we assume that there are fixed initial stocks of the state variables (non-renewable resource and the capital stock), and that we specify that the state variables are equal to zero at the end of the optimisation period. We also assume that a backstop technology exists that will replace oil as a productive input at the end (terminal point) of the optimisation period, t = T. This explains why we set ST = 0, as there is no point having any stocks of oil remaining once the backstop technology has replaced oil in production. We assume that the capital stock, Kt, associated with the oil input, will be useless for the backstop techno-logy, and therefore will be consumed completely by the end of the optimisation period, so KT = 0. Implicitly in this simulation, we assume that a new capital stock, appropriate for the backstop technology, will be accumulated out of the resources available for consumption. So Ct in this model should be interpreted as consumption plus new additions to the (backstop) capital stock. The question of how much should be saved to accumulate this new capital stock is beyond the scope of our simple model. As the notation will have made clear, this is a discrete-time model. We choose each period to be 10 years, and consider a 10-period (t = 0, 1,. . ., 9) time horizon of 100 years beginning in 1990 (t = 0). The following data are used in the simulation: Estimated world oil reserve =11.5 (units of 100 billion barrels) World capital stock = 4.913 (units of 10 trillion $US) Efficiency parameter A = 3.968 Utility discount rate = 5% The value of the efficiency parameter is estimated under the assumption that world aggregate output over the 1980s was $US179.3 trillion, and aggregate oil extraction was 212.7 billion barrels. We cannot obtain an analytical solution for this problem. But it is possible to solve the problem numerically on a computer. Table 14.1 and Figures 14.614.8 show the numerical solution. Figure 14.6 shows that consumption rises exponentially through the whole of the optimisation period (the figure only shows the first part of that period, from 1990 up to 2080); output (Ft) also rises continuously, but its growth rate slows down towards the end of the period shown in the figure; this arises because at the terminal point (not shown in these figures), the capital stock has to fall to zero for the reason indicated above. Table 14.1 Numerical solution to the oil extraction and optimal consumption problem Welfare (p.v.) = 46.67668TimeCtQ(Kt, Rt)KtRtStq(t+ + 1)P(t + 1)"Q/"K(t)1990s3.734218.05184.91302.277011.50000.26780.21233.30692000s13.681960.834719.23061.99479.22300.07310.22292.84712010s45.3301182.835366.38341.72327.22830.02210.23412.47882020s137.2777493.8224203.88861.46375.50510.00730.24582.17982030s383.60601204.3708560.43341.21674.04130.00260.25811.93412040s997.33502654.79921381.19820.98242.82470.00100.27101.72992050s2430.10805261.71983038.66240.76111.84230.00040.28451.55842060s5584.89709217.11255870.27420.55251.08120.00020.29871.41312070s12174.562113608.65789502.48970.35640.52870.00010.31371.28892080s25298.482514361.897110936.58540.17240.17240.00000.32931.18192090s0.00.00000.00000.00000.3548 10 trillion US$ 100 billion barrels% Growth rates:2000266.3899237.0006291.422212.401219.800474.00645.00002010231.3150200.5443245.197313.607121.627171.25455.00002020202.8399170.0914207.137815.060723.840268.55175.00002030179.4380143.8874174.872316.878326.588565.91805.00002040159.9894120.4304146.451819.253030.105363.36855.00002050143.660298.1965120.001922.532034.779760.91365.00002060129.820975.173093.186127.408141.311058.55995.00002070117.990847.645661.8747 35.4951 51.0972 56.31105.00002080107.79795.535015.0918 51.3311 67.3996 54.16795.0000In Figure 14.7 we observe the shadow price growing over time at the rate , and so satisfying Hotelling s rule; the shadow price of capital falls continuously through time. In Figure 14.8, we see that the oil stock falls gradually from its initial level towards zero; note that as the shadow price of oil rises over time, so the rate of extraction falls towards zero. Not surprisingly, the optimal solution will require that the rate of extraction goes to zero at exactly the same point in time, tf, that the stock is completely exhausted. Note that within 100 years, the oil stock has fallen to a level not significantly different from zero; it is optimal to deplete the stock of oil fairly rapidly in this model. What happens after the year 2090? You should now try to deduce the answer to this question. Summary In this chapter, we have constructed a simple economic model of optimal resource depletion, and studied several of its variants. The framework we use is a version of the so-called optimal growth models, built around a production function in which natural resources are inputs into the production process. The solution to an optimal resource depletion model should be one in which a set of static and intertemporal efficiency conditions as discussed in Chapters 5 and 11 are satisfied. In addition, optimality requires that, of the many possible efficient depletion paths that may be available, the one chosen is that which maximises the relevant objective function (in this case the intertemporal social welfare function). The characteristics of a socially optimal pattern of natural resource use over time will depend on the particular nature of the social welfare function that is deemed to be appropriate. In this chapter as generally throughout the text the special case chosen is that of the utilitarian social welfare function, in which future utility is discounted at the positive rate . Given that the objective function used here is specified in terms of units of utility, all prices and values referred to in this chapter are also specified in units of utility. This poses a practical problem in that utility is an unobservable variable. However, given the form of the utility function U = U(C), all results could be stated equivalently in terms of units of consumption (or income). One efficiency property to which resource economists pay considerable attention is the so-called Hotelling rule. This intertemporal efficiency condition requires that the real rate of return to a resource owner should equal the social discount rate. For a non-renewable resource available in known, fixed quantity, the Hotelling rule implies that the net price of the resource, specified in utility units, should grow at the proportionate rate . The Hotelling rule has general applicability and does not apply only to non-renewable resources. We briefly state how the rule applies to renewable resources. This will be examined at length in Chapter 17. Although we did not discuss it in this chapter, the rule also applies where resource extraction or use generates adverse external effects. This case is examined at length in Chapter 16. Further reading The mathematics underlying our analyses is presented simply and clearly in Chiang (1984, 1992). Kamien and Schwartz (1981) is also an excellent reference for optimal control theory. Excellent advanced-level presentations of the theory of efficient and optimal resource depletion can be found in Baumol and Oates (1988), Dasgupta and Heal (1979), and Heal (1981). Kolstad and Krautkraemer (1993) is particularly insightful and relatively straightforward. Dasgupta and Heal (1974) is also a comprehensive study, and is contained along with several other useful (but difficult) papers in the May 1974 special issue of the Review of Economic Studies. Less difficult presentations are given in Hartwick and Olewiler (1998), Anderson (1985, 1991), and the Fisher and Peterson survey article in the March 1976 edition of the Journal of Economic Literature. For an application of this theory to the greenhouse effect, see Barbier (1989b) or Nordhaus (1982, 1991a). Barbier (1989a) provides a critique of the conventional theory of resource depletion, and other critical discussions are found in Common (1995) and Common and Perrings (1992). Discussion questions Are non-renewable resources becoming more or less substitutable by other productive inputs with the passage of time? What are the possible implications for efficient resource use of the elasticity of substitution between non-renewable resources and other inputs becoming (a) higher, and (b) lower with the passage of time? 2. Discuss the possible effects of technical progress on resource substitutability. 3. Recycling of non-renewable resources can relax the constraints imposed by finiteness of non-renewable resources. What determines the efficient amount of recycling for any particular economy? Problems Using the relationship EMBED Unknown demonstrate that if the utility function is of the special form U(C) = C, the consumption rate of discount (r) and the utility rate of discount are identical. Using equation 14.15 in the text (that is, the Hotelling efficiency condition), demonstrate the consequences for the efficient extraction of a non-renewable resource of an increase in the social discount rate, . The simplest model of optimal resource depletion is the so-called  cake-eating problem in which welfare is a discounted integral of utility, utility is a function of consumption, and consumption is equal to the amount of the (non-renewable) resource extracted. That is: EMBED Unknown (a) Obtain the Hamiltonian, and the necessary first-order conditions for a welfare maximum. (b) Interpret the first-order conditions. (c) What happens to consumption along the optimal path? (d) What is the effect of an increase in the discount rate? Appendix 14.1 The optimal control problem and its solution using the maximum principle Optimal control theory, using the maximum principle, is a technique for solving constrained dynamic optimisation problems. In this appendix we aim to explain what is meant by a constrained dynamic optimisation problem; show one technique the maximum principle by which such a problem can be solved. We will not give any proofs of the conditions used in the maximum principle. Our emphasis is on explaining the technique and showing how to use it. For the reader who wishes to go through these proofs, some recommendations for further reading are given above. After you have finished reading this appendix, it will be useful to go through Append-ices 14.2 and 14.3. Appendix 14.2 shows how the maximum principle is used to derive the optimal solution to the simple non-renewable resource depletion problem discussed in Part 1 of this chapter. Appendix 14.3 considers the optimal allocation of a renewable or non-renewable resource in the case where extraction of the resource involves costs the model discussed in Part 2 of this chapter. Let us begin by laying out the various elements of a constrained dynamic optimisation problem. In doing this, you will find it useful to refer to Tables 14.2 and 14.3, where we have summarised the key elements of the optimal control problem and its solution. Table 14.2 The optimal control problem and its solution7 Objective function (see note 1)EMBED UnknownSystemEMBED Unknownx(t0) = x0Terminal statex(tT) = xTx(tT) freeTerminal pointtT fixedtT freetT fixedtT freeHamiltonianH = H(x, u, t, () = L(x, u, t) + (f(x, u, t)Equations of motionEMBED UnknownMax H (see note 2)EMBED UnknownTransversality conditionx(tt )=xTEMBED UnknownEMBED UnknownEMBED UnknownNotes to Table 14.2 1. The term F(x(tT)) may not be present in the objective function, and cannot be if tT = (. 2 The Max H condition given in the table is for the special case of an interior solution (u is an interior point). A more general statement of this condition (the maximum principle) is: u(t) maximises H over u(t) = U, for 0 d" t d" tT. Table 14.3 The optimal control problem with a discounting factor and its solution8 Objective function (see note 1)EMBED UnknownSystemEMBED Unknownx(t0) = x0Terminal statex(tT) = xTX(tT) freeTerminal pointtT fixedtT freetT fixedtT freePresent-value HamiltonianH = H(x, u, t, () = L(x, u, t) + (f(x, u, t)Current-value HamiltonianEMBED UnknownEquations of motionEMBED UnknownEMBED UnknownMax Hc (see note 2)EMBED UnknownTransversality conditionx(tt ) = xTEMBED UnknownEMBED UnknownEMBED UnknownNotes to Table 14.3 Note 1: The term F(x(tT))e-T may not be present in the objective function, and cannot be if tT = (. Note 2: two versions of the Hamiltonian function are given. The first is known as the  present-value Hamiltonian, as the presence of the term e-t in the objective function (which converts magnitudes into present-value terms) carries over into the Hamiltonian, HP. The second is known as the current-value Hamiltonian (see Chiang, 1992, p. 210). Premultiplying HP by et removes the discounting factor from the expression, and hence HC is expressed in current-value terms. For mathematical convenience, it is usually better to work with the Hamiltonian in current-value terms. Note 3: The Max HC condition given in the table is for the special case of an interior solution (u is an interior set). A more general statement of this condition (the maximum principle) is: u(t) maximises HC over u(t) ( U, for 0 d" t d" tT. Note 4: The term "F/"T does not enter the transversality condition in the final line of the table if F() = 0. The function to be maximised is known as the objective function, denoted J(u). This takes the form of an integral over a time period from an initial time t0 to the terminal time tT. Two points should be borne in mind about the terminal point in time, tT: In some optimisation problems tT is fixed; in others it is free (and so becomes an endogenous variable, the value of which is solved for as part of the optimisation exercise). In some optimisation problems the terminal point is a finite quantity (it is a finite number of time periods later than the initial time); in others, the terminal point is infinite (tT = (). When a problem has an infinite terminal point in time, tT should be regarded as free. 2. The objective function will, in general, contain as its arguments three types of variable: x(t), a vector of n state variables at time t; u(t), a vector of m control (or instrument) variables at time t; t, time itself. Although the objective function may contain each of these three types of variables, it is not necessary that all be present in the objective function (as you will see from the examples worked through in Appendices 14.2 and 14.3). 3. The objective function may (but will often not) be augmented with the addition of a final function, denoted by the function F() in Tables 14.2 and 14.3. Its role (where it appears) will be explained below. (The applications in Chapter 14 did not involve the use of a final function.) 4. The solution to a dynamic optimal control problem will contain, among other things, the values of the state and control variables at each point in time over the period from t0 to tT. It is this that makes the exercise a dynamic optimisation exercise. 5. Underlying the optimal control problem will be some economic, biological or physical system (which we shall call simply the economic system), describing the initial values of a set of state variables of interest, and how they evolve over time. The evolution of the state variables over time will, in general, be described by a set of differential equations (known as state equations) of the form: EMBED Unknown where EMBED Unknown is the time derivative of x (the rate of change of x with respect to time). Note that as x is a vector of n state variables, there will in general be n state equations. Any solution to the optimal control problem must satisfy these state equations. This is one reason why we use the phrase constrained dynamic optimisation problems. 6. A second way in which constraints may enter the problem is through the terminal conditions of the problem. There are two aspects here: one concerns the value of the state variables at the terminal point in time, the other concerns the terminal point in time itself. First, in some problems the values that the state variables take at the terminal point in time are fixed; in others these values are free (and so are endogenously determined in the optimisation exercise). Secondly, either the particular problem that we are dealing with will fix the terminal point in time, or that point will be free (and so, again, be determined endogenously in the optimisation exercise). 7. The optimisation exercise must satisfy a so-called transversality condition. The particular transversality condition that must be satisfied in any particular problem will depend upon which of the four possibilities outlined in (6) applies. (Four possibilities exist because for each of the two possibilities for the terminal values of the state variables there are two possibilities for the terminal point in time.) It follows from this that when we read Tables 14.2 and 14.3, then (ignoring the column of labels) there are four columns referring to these four possibilities. Where cells are merged and so cover more than one column, the condition shown refers to all the possibilities it covers. We shall come back to the transversality condition in a moment. 8. The control variables (or instruments) are variables whose value can be chosen by the decision maker in order to steer the evolution of the state variables over time in a desired manner. 9. In addition to the three kinds of variables we have discussed so far time, state and control variables a fourth type of variable enters optimal control problems. This is the vector of co-state variables ( (or ( in the case where the objective function contains a discount factor). Co-state variables are similar to the Lagrange multiplier variables one finds in static constrained optimisation exercises. But in the present context, where we are dealing with a dynamic optimisation problem over some sequence of time periods, the value taken by each co-state variable will in general vary over time, and so it is appropriate to denote ( (t) as the vector of co-state variables at time t. 10. The analogy of co-state variables with Lagrange multipliers carries over to their economic interpretation: the co-state variables can be interpreted as shadow prices, which denote the decision makers marginal valuation of the state variable at each point in time (along the optimal time path of the state and control variables). 11. Finally, let us return to the transversality condition. Looking at the final rows in Tables 14.2 and 14.3 you will see four possible configurations of transversality condition. All relate to something that must be satisfied by the solution at the terminal point in time. Where the terminal value of the state variables is fixed, this will always be reflected in the transversality condition. On the other hand, where the terminal value of the state variables is free, the transversality condition will usually9 require that the shadow price of the state variables be zero. Intuitively, this means that if we do not put any constraints on how large the stocks of the state variables must be at the terminal point in time, then they must have a zero value at that time. For if they had any positive value, it would have been optimal to deplete them further prior to the end of the planning horizon. Note also that whenever the terminal point in time is free (whether or not the state variables are fixed at the terminal point), an additional part of the transversality condition requires that the Hamiltonian have a zero value at the endogenously determined terminal point in time.10 If it did not, then the terminal point could not have been an optimal one! The general case referred to in Tables 14.2 and 14.3, and special cases In the description we have given above of the optimal control problem, we have been considering a general case. For example, we allow there to be n state variables and m control variables. In some special cases, m and n may each be one, so there is only one state and one control variable. Also, we have written the state equation for the economic system of interest as being a function of three types of variables: time, state and control. In many particular problems, not all three types of variables will be present as arguments in the state equation. For example, in many problems, time does not enter explicitly in the state equation. A similar comment applies to the objective function: while in general it is a function of three types of variables, not all three will enter in some problems. Finally, often the objective function will not be augmented by the presence of a final function. Limitations to the optimal control technique outlined in this appendix The statement of the optimal control problem and its solution given in this appendix is not as general as it might be. For example, the terminal condition might require that a control variable must be greater than some particular quantity (but is otherwise unconstrained). As neither this (nor any other) complication arises in the examples discussed in this book, we do not go through them here. Details can be found in Chiang (1992). The presence of a discount factor in the objective function For some dynamic optimisation problems, the objective function to be maximised, J(u), will be an integral over time of some function of time, state variables and control variables. That is: EMBED Unknown However, in many dynamic optimisation problems that are of interest to economists, the objective function will be a discounted (or present-value) integral of the form: EMBED Unknown For example, equation 14.8 in the text of this chapter is of this form. There, L is actually a utility function U() (which is a function of only one control variable, C). Indeed, throughout this book, the objective functions with which we deal are almost always discounted or present-value integral functions. The solution of the optimal control problem The nature of the solution to the optimal control problem will differ depending on whether or not the objective function contains a discounting factor. Table 14.2 states formally the optimal control problem and its solution using general notation, for the case where the objective function does not include a discount factor. Table 14.3 presents the same information for the case where the objective function is a discounted (or present-value) integral. Some (brief) explanation and discussion of how the conditions listed in Tables 14.2 and 14.3 may be used to obtain the required solution is provided below those tables. However, we strongly urge you also to read Appendices 14.2 and 14.3, so that you can get a feel for how the general results we have described here can be used in practice (and how we have used them in this chapter). Interpreting the two tables It will help to focus on one case: we will look at an optimal control problem with a discounting factor, an infinite time horizon (so that tT is deemed to be free), and no restriction being placed on the values of the state variable in the terminal time period (so that x(tT) is free. The relevant statement of the optimal control problem and its solution is, therefore, that given in the final column to the right in Table 14.3. We can express the problem as EMBED Unknown subject to EMBED Unknown and x(t0) = x0, x(0) given, x(tT) free. To obtain the solution we first construct the current-value Hamiltonian: HC = L(x, u, t) + (f(x, u, t) The current-value Hamiltonian consists of two components: The first L(x, u, t) is the function which, after being multiplied by the discounting factor and then being integrated over the relevant time horizon, enters the objective function. Note carefully by examining Table 14.3 that in the Hamiltonian the L function itself enters, not its integral. Furthermore, although the discounting factor enters the objective function, it does not enter in the current-value Hamiltonian. The second component that enters the Hamiltonian is the right-hand side of the state variable equations of motion, f(x, u, t), after having been premultiplied by the co-state variable vector in current-value form. Remember that in the general case there are n state variables, and so n co-state variables, one for each state equation. In order for this multiplication to be conformable, it is actually the transpose of the co-state vector ( that premultiplies the vector of functions from the state equations. Our next task is to find the values of the control variables u which maximise the current-value Hamiltonian at each point in time; it is this which gives this approach its name of the maximum principle. If the Hamiltonian function HC is non-linear and differentiable in the control variables u, then the problem will have an interior solution, which can be found easily. This is done by differentiating HC with respect to u and setting the derivatives equal to zero. Hence in this case one of the necessary conditions for the solution will be & vw)*GH    &(*,.<>XZ\^X ſԿ{jwhN|hYEHUjWI h$xUVjh$xUjhrhYEHUjVI h$xEHUVjh$xEHU h$xEHh$x6@] h$x0Jd h$x6] h$x@hrh$x6]h$x56\] h$x5\h$x0&1d&  & * pv",Y"4"#""`G"w"d; "HS"`G"""`G"| "Y" " " "^0"| "| "&" "0_B @gdr Oh^h`? & F'gdhe>gdr+>$a$>=@$gd0~MRU#>,? <` "%'@(,,1182==ACC0EGiK""" """@/"| "&"!"Y"!""| "!"T"T)";"0""L"45"@/" "Y"@/" : 0^`0gd0~MBB @gdrX Y j k n o p q s t !!~!".(0(:(<(,,,,,,,,,---... .n.p.r.z.|.~.....11%1jhlVh$xUhlVh$x0JdhlVh$x@hlVh$x6EH]hlVh$x6] hlVh$xjNhN|hYEHUjXI h$xUVjh$xUh$x6@] h$x@h$x h$x6]2%1&1'1(1144CC"L$L\L^LLLLLLLDPFP~PPPPPPPPdehiGiNijjll}mmtttwθέθΕ΄yhN|;@CJaJhN|jkhJhYEHUj]I h$xUVjhN|hYEHUj\I h$xUVjh$xU h$x6] jh$xh$x hlVh$xjhlVh$xUjk hlVhYEHUj[I hlVh$xUV,iKLLNPPRT0^NcdefgJhllGo!rttv8wOxx" "" " "" "|"@>"^0"Y" "@/"" " :" "!"&"!" "p"p" "8% & F0^`0gdN|gdN|&gdN| Oh^h`Bwwwwwwwwwwzz{{{{V{X{r{t{v{x{KLRSVWYZ[]^_abcklm|}~ʽ~h$x6@EH]h$x6@]h$x6EH]h$x6@] h$xEHh$xCJaJ h$x6]h$xj)h.hYEHUj_I hN|UVjDh.hYEHUj^I hN|UVjhN|UhN| hN|6]1xx8yz{V{z{ pv 5vŌ ؍N"8"8"p"e"8"d""M"P"C" " "@"" """ "@"@"" "& P & Fh^h`$$a$gdN|gdN| gdN|% & F0^`0gdN|NBwLėRvК*j)C<ͩw" "@"`"!"L"" """"T"""l;"p""(Y "p""","Y"" ""TA "B @gdB @gdr 0^`0gd0~MКҚښܚޚ*,!")*,457 jkxyzjc hhYEHUjaI h$xUVj`hhYEHUj`I h$xUVjh$xUh$x6@]h$x;@CJaJh$x6EH] hh$x h$xEH h$x0Jd h$x6]h$x7z{z{¢âǢȢɢv \])*789:C󩞩ujfI hlVh$xUVj*hlVhYEHUjdI hlVh$xUVjhlVh$xU hlVh$xj'hYEHUjcI h$xUVj/$hhYEHUjbI h$xUVh$x6EH] h$xEH h$x6]h$xjh$xU.89:<©éĩwxy|}~ƧƖƖƖƖƖƖƲƂuƖƖƖjz3hhYEHUjjI h$xUVh$x6EH] h$x6]j0hYEHUjiI h$xUVjh$xU h$xEH h$x0Jdh$xhlVh$x6EH]hlVh$x6] hlVh$xjhlVh$xUj-hlVhYEHU-"=OSe?q°ݰȴ`5`i"@/"<"l;"@""@""&" "@"0"0""l;" """Y" "P)"p"P)""` "Y & Oh^h`B gdheB !=>KLMNOSTabcde?qrŸšxmcjChYEHUjoI h$xUVj@hmhhYEHUjnI h$xUV h$x6]j=hYEHUjmI h$xEHUVj:hYEHUjlI h$xEHUV h$xEHjh$xEHUj6hhYEHUjkI h$xUVjh$xUh$x jh$x&°ðаѰҰӰ $&*,ܴ޴|~\^``adefgopۼۼۼۼۼۼۼ۳ۼ۳ۼۦ۠ۏۼۼۆۼh$x6@]h$x;@CJaJ h$xEH h$x0Jd jh$x h$xEHh$x6EH] h$x6]j JhmhhYEHUjqI h$xUVh$xjh$xUjFhmhhYEHUjpI h$xUV4fgijOV^`lnprt"$df*+67DEjrI hlVh$xUVjhlVh$xUhlVh$x6] hlVh$xh$x6@] h$xEHh$x6EH]h$x6@]h$x;@CJaJh$xCJaJh$x h$x6]9A;az:^~j2]p7Q:"w"M"" "!" " "d; "Y"<#"|p("d; "p"# " " " "# "@/""@""d"0"@+"YB @gdrB+ &dP#EFG:<VXZ\~jkxyz{&'()]^_bcdefgɼԵԪԒ|sjbYPh$x6EH] jrh$xEHh$x@EHh$x6@]h$x6EH]h$x6@]jWhhYEHUjuI h$xUVjoThhYEHUjtI h$xUV h$x6]jPhhYEHUjsI h$xUVh$xjh$xU hlVh$xjhlVh$xUjMhlVhYEHUgpqrs+,78EFGH_e` "$&Ÿ h$x@jbhhYEHUjxI h$xUVjr^hhYEHUjwI h$xUVjZhhYEHUjvI h$xUVjh$xU jrh$xEHh$x6EH] h$x6]h$x;@CJaJh$x0`(^6p=#;" "@" " """@"@"@""@" "@"@" "&";""^0""L""p"B @gdr `^``gd0~M P & Fh^h`B()lmwx}~BC =>UVmn?@Aqr BDnpr#$%*+78Ƽ j h$xj fhYEHUjyI h$xUVjh$xU jGh$x h0~M@h$x6EH] h$x6]h$x h$xEHG^L^y[      e b.@dxl"4:U"L"@"<"l;"@""@""h""""""Y"Y""0_"P"Y"""" B gd0~MB gdheB 0^`0gd0~M89LMZ[\]yzR S V W X          Ÿyljsh-OhYEHUj}I h$xUV jGh$x j h$x h$x6]johYEHUj|I h$xUVjlhYEHUj{I h$xEHUV h$xEHjh$xEHUjhh-OhYEHUjzI h$xUVjh$xUh$x jh$x(                bdfnprt~+,XY@۸ۤۛ۔ۛ۔۔ۛۇۇ~ۤۤ jGh$x@ jwh$x h$x@ h$x;@h$x6EH] h$x6]js|h0hYEHUjI h$xUVj0yh-OhYEHUjI h$xUVh$xjh$xUj vh0hYEHUj~I h$xUV1@B\^`bz|~\^`bdf12?@ABL պ󕋕rej?hlVhYEHUjI hlVh$xUVjhlVh$xUhlVh$x6] hlVh$xjh0hYEHUjI h$xUV h$x@h$x6@EH] h$x6]h$x6EH] h$x@jh0hYEHUjI h$xUVh$xjh$xU$l1L,!w"%%'())**c-/"000N02 388{;;;U<" ""Dd<""@+"| ""(Y """""h#"F$""@""";"h#"| "!"@"@" P & Fh^h`B @gdrBB gdhe    ,! " "&&''(((((((((((((((())))))*********ջΨΨΗՌwjΨjh0hYEHUjI h$xUVjyhYEHUjI h$xUVh$x6@EH] h$xEHh$x6EH]jӈhYEHUjI h$xUVjh$xU h$x6]h$x hlVh$xhlVh$x@hlVh$x6@EH]hlVh$x@)*****3,4,5,c-//////000 00080:0<00001212222222X7Z7z7|7~788I8J8K8FFGG#H$H%H(H)H*H+H,H0H1H3H̿h$x6@]h$x6EH]h$x;@CJaJ h$x@jɒhlVhYEHUjI hlVh$xUVjhlVh$xU hlVh$xh$x6EH] h$x6]h$x;U<T=u>xBOD-EyEaGyGGGG#HiHHI/IAISIaIsIIIII"""45"@/""L"@/";";";"P5"8"")"؂"8"t;"؂"8"8S"8S" "̈" %gd0~M & 0^`0gd0~M P & Fh^h`3H4H6H7H8HI?I@IAIBIOIPIQIRIaIbIjwh0hYEHUjI h$xUVjh0hYEHUjI h$xUVjh0hYEHUjI h$xUVjh$xUh$x6@EH]h$x6@]h$x6EH] h$x6]h$x h$xEH3bIoIpIqIrIsIyIzIIIIIIIIIIIIIIIIIIIIIIIIIIռՌtj_hlVh$xCJaJjhYEHUjI h$xUVjԬhlVhYEHUjI hlVh$xUVjۨhYEHUjI h$xUVjWhlVhYEHUjI hlVh$xUVjhlVh$xU hlVh$xjh$xUjeh0hYEHUjI h$xUVh$x IIIIIIIIIIIIIJJBJHJIJaJjJkJxJyJzJ{J|J}J~JJ德si\Sh0~Mh$xEHjh0~Mh$xEHUh0~Mh0~M6]jhlVhYEHUjI hlVh$xUVjhlVh$xU hlVh$x h0h$x h0~Mh$xh$x6EH] h$x6]h$x!jhlVhYCJEHUaJ#jI hlVh$xCJUVaJhlVh$xCJaJjhlVh$xCJUaJIIBJaJ|JJJJ,KbKZLLMMM~OOORTUVV"8S"TM"|""k" "hA"""",Y"5"@"T"h" "k"""""0_B !&dP !&dPgdN &dPgd0~M &dPJJJJJJJJJJJJJJJJJJJJJJJKKKKK KKKK K,K-K/K1K3K4K5K8K9K:K\L\Z\j\x\\\C؈ ؈ ؈ ؈؈P؈؈؈؈ _$$Ifa$_$If \\ ؈kd$$Ifl  El$")'0)$$$$4 la\\\\\\\\ ]]C؈ ؈ ؈ ؈؈P؈؈؈؈ _$$Ifa$_$If ]] ؈kd$$Ifl  El$")'0)$$$$4 la]&]6]H]X]f]t]]]]C؈ ؈ ؈ ؈؈P؈؈؈؈ _$$Ifa$_$If ]] ؈kd$$Ifl  El$")'0)$$$$4 la]]]]]]]^ ^^C؈ ؈ ؈ ؈؈P؈؈؈؈ _$$Ifa$_$If ^^ ؈kd$$Ifl  El$")'0)$$$$4 la^^$^.^7^>^E^L^S^Z^C؈ ؈ ؈ ؈؈P؈؈؈؈ _$$Ifa$_$If Z^[^ ؈kd$$Ifl  El$")'0)$$$$4 la[^a^j^t^~^^^^^^C؈ ؈ ؈ ؈؈P؈؈؈؈ _$$Ifa$_$If ^^ ؈kd$$Ifl  El$")'0)$$$$4 la^^^^^^^^^^C؈ ؈ ؈ ؈؈P؈؈؈؈ _$$Ifa$_$If ^^ ؈kd$$Ifl  El$")'0)$$$$4 la^^^____#_*_1_C؈ ؈ ؈ ؈؈P؈؈؈؈ _$$Ifa$_$If 1_2_ ؈kdv$$Ifl  El$")'0)$$$$4 la2_8_C_N_X___f_m_t_{_C؈ ؈ ؈ ؈؈P؈؈؈؈ _$$Ifa$_$If {_|_ ؈kdj$$Ifl  El$")'0)$$$$4 la|__________C؈ ؈ ؈ ؈؈P؈؈؈؈ _$$Ifa$_$If __ ؈kd^$$Ifl  El$")'0)$$$$4 la__________C؈ ؈ ؈ ؈؈P؈؈؈؈ _$$Ifa$_$If __ ؈kdR$$Ifl  El$")'0)$$$$4 la__`````C؈؈؈؈؈؈ _$$Ifa$ _$$Ifa$``.`C؈=(؈_$IfkdF$$Ifl4ֈl$")0)4 laf4.`/`4`=`F`O`X`a`j`q`r`؈C؈ ؈ ؈ ؈؈l؈؈؈؈ _$$Ifa$_$If\kd$$Ifl4"))0)4 laf4 r`s` ؈kd$$Ifl  (l$")D0)$$$$4 las`x`````````C؈ ؈ ؈ ؈؈l؈؈؈؈ _$$Ifa$_$If `` ؈kd$$Ifl  (l$")D0)$$$$4 la``````````C؈ ؈ ؈ ؈؈l؈؈؈؈ _$$Ifa$_$If `` ؈kd$$Ifl  (l$")D0)$$$$4 la`a aaa$a-a6a=a>aC؈ ؈ ؈ ؈؈l؈؈؈؈ _$$Ifa$_$If >a?a ؈kdx$$Ifl  (l$")D0)$$$$4 la?aDaMaVa_ahaqazaaaC؈ ؈ ؈ ؈؈l؈؈؈؈ _$$Ifa$_$If aa ؈kdl$$Ifl  (l$")D0)$$$$4 laaaaaaaaaaaC؈ ؈ ؈ ؈؈l؈؈؈؈ _$$Ifa$_$If aa ؈kd`$$Ifl  (l$")D0)$$$$4 laaaaaaaaa bbC؈ ؈ ؈ ؈؈l؈؈؈؈ _$$Ifa$_$If bb ؈kdT$$Ifl  (l$")D0)$$$$4 labb,bnZutu!v?vyyyyyyxzzz|z~zzzzz~~~~~~/01QR_`폆{jI h$xUVh$x6EH]h$x0Jd6] h$x;@jUhYEHUjI h$xUVj>hYEHUjI h$xUVjh$xUh$x6@]h$xCJaJh$x6@EH] h$x6]h$xh$xCJaJ/cgghjlnoprrvu\wqwxxxxyyy"F'" " "M"p"|" "p"P"| "!""| ""@"@"@"@""| P & F ^` P & F^` P & Fh^h`` &dP&yyy6{|~~~Ox1Qc"@" " """D"@"@"@"@"L" "@"@"&""   _$$Ifa$_$Ifa Oh^h` P & F^`B P & Fh^h` P & Fh^hB`P & Fh^h`gdrk`abklyz{|}~DžȅɅυЅх؅مڅ潫Ğ潫潫潫潫潫 jlh$xhNh$x6EHH*]h$x6EH]h$x5@ \ h$x6] h$x5\ h$xEHjhYEHUjI h$xUVh$xjh$xUjjhNhYEHU;cdk} ؈{o {  _$$Ifa$_$IftkdN$$Ifl40f04 laf4w q؈ho h  _$$Ifa$_$Ifkd$$Ifl4FfF j 0    4 laf4Džυ؅wq؈qqq@qz_$Ifkd$$Ifl4FfF j 0    4 laf4)TN؈E$E$ _$$Ifa$_$Ifkd$$Iflrf R04 la !#$&'>?LMNOPQWXfgtuvwxf"jhrkhY5EHH*U\jI h$x5H*UV\jh$x5H*U\h$x5H*\ h$xH*jhYEHUjI h$xUVj hYEHUjhYEHUjI h$xUVjh$xUh$x6@] jlh$x h$x6] h$x5\h$x()*>PRH `{ {ze _$$Ifa$_$Iftkds$$Ifl40f04 laf4RSfxwe qHhe _$$Ifa$_$Ifkd$$Ifl4Ff^R0    4 laf4xye `{o { _$$Ifa$_$Iftkd$$Ifl40f04 laf4†Ćņ҆ӆԆՆ?@ADEQR  $&2468@BVX^`bz| 翵箞箞箞献h$x0Jd6]h$x;@CJaJ jh$xh$x6EH] h$x5\ h$x6]j2hYEHUjI h$xUVj]hYEHUjI h$xUVjh$xUh$xh$x5H*\jh$x5H*U\3ÆĆֆw q؈q؈q q@؈qz _$Ifkd$$Ifl4FfF j 0    4 laf4ֆ׆Gf LT R"؈E"ԈE" C" =`_$Ifa bp^`pgdrk_kd $$Iflrf0 |R04 la  LNhjlnډ܉މ$%&-./PQTUVWYZ\]_뻴h$x6EH]h$x5@ \ h$x6] h$x5\ h$xEHjhYEHUjI h$xUVj hYEHUjI h$xUVjh$xUh$xh$x6EH];Lpr {؈ C _$Iftkd$$Ifl40 n0!4 laf4 _$$Ifa$ډw q؈h hC  _$$Ifa$_$Ifkdn$$Ifl4F n~  0!    4 laf4$-5wq؈qq@q@q+_$Ifkdb$$Ifl4F n~  0!    4 laf456PbTN`E$E$ _$$Ifa$_$Ifkd$$Iflrh  nf0!4 la_`rstuwxz{͊Ίۊ܊݊ފߊŸ~sjI h$xUVh$x6H*]jhYEHUjI h$xUVjhYEHUjI h$xUVjh$xUjhY6EHU]jI h$x6UV]jh$x6U]h$x6@] h$x5\ h$x6]h$x jlh$x,H `{ _$$Ifa$_$Iftkd$$Ifl40 n0!4 laf4͊ߊ``{ {+e _$$Ifa$_$Iftkd/$$Ifl40 n0!4 laf4we qHhe _$$Ifa$_$Ifkd!$$Ifl4F n0!    4 laf42357;<=>?LMNOQSTabcdfgtuvwձ锇|ohh_h$x6EH] h$x6]j-h0i+hYEHUjI h$xUVjW*h0i+hYEHUjI h$xUV"j%h0i+hY5EHH*U\jI h$x5H*UV\jh$x5H*U\ h_h$xh$x5H*\ h$xH* h$x5\h$xjh$xUj"hYEHU%2>Pe `{ {C _$$Ifa$_$Iftkd$$$Ifl40 n0!4 laf4PQRSefxw q؈qW؈qx q@؈q+_ _$Ifkd)$$Ifl4F n~  0!    4 laf4xyxT_ R"HR"PR"|R"Ĉ R"0E"P P & Fh^h`bkd1$$Iflr0  n.P0!4 lahjlvzMNOP\]^defgijkl 8:>@ޒ+,-tuvh$x6EH] h$xEH jh$x h$x5\ h$xEH h$xEH jh$xh$x6EH] h$x6]h$x6EH] h$xH*h$xDޕߕ ǖȖʖ˖̖͖Җؖܖݖ   әԙՙٙڙۙ›Û윲̛ٛڛۛܛŻj4hlVhYEHUjI hlVh$xUVjhlVh$xU hlVh$xj1hYEHUjI h$xUVjh$xU h$x5\ jh$x h$x6]h$x h$xEHh$x6EH]6x(=ʖ #!ś,9џ͢B}Ŭ"TA ""@"("("@""Y""@/"8"4;""""&" "P'"Y"C" B P & Fh^h` P & Fh^h O]^`P & F h^`gd_ O^`67GHst,]^cd ?@/1}ŬWXmnGB~βϲвѲ<=JKLMᦜᑇj;hYEHUjI h$xUVj7hYEHUjI h$xUVjh$xU h$x5\ h$x0Jd h$x6] jmh$x5\ jlh$x5\h$xhlVh$x6] hlVh$xhlVh$x5\5ŬGB~<N?kθ|:X75VhD"^0" "@/" "h "\ " "\ "Y" "@+" "X"@"\ "@""@"DS"@"(;"|"R""|#" " Oh^h`BWXxy?k;<θYZ[ܹݹ޹߹źƺǺȺͺκϺкѺպֺ׺ٺں:;<?@ABDEGHLɼ h$xEH h$x5\jBhYEHUjI h$xUVj>h_hYEHUjI h$xUVh$x6EH] h$x@h$x6@] h$x6]h$xjh$xU@BD\^xz|~ֿؿڿܿ  ¸vh\hlVh$x6@]hlVh$x6@EH] jmhlVh$xhlVh$x6EH] hlVh$xhh$x6EHH*]hlVh$x6@EH]hlVh$x6]jRhYEHUjI h$xEHUV h$xEHjh$xEHUjh$xUjNhYEHUjI h$xUVh$x$"$&.0248Z\^`bfhj26^`zjhlVh$xUhlVh$x@hlVh$x6@ EH]hlVh$x6EH]hlVh$x6@EH]hhlVh$x6@EH] jmhlVh$x hlVh$xhlVh$x6EH]hlVh$x6]1z|~z|X24NPRTlnbķҳyodZjahYEHUjI h$xUVj^hYEHUjI h$xEHUV h$xEHjh$xEHUj[hYEHUjI h$xUVjh$xUh$xjnXhlVhYEHUjI hlVh$xUV hlVh$xjhlVh$xUjbUhlVhYEHUjI hlVh$xUVbdfhlnpxz|HJdӻӣә|xqfxq]xUxjh$xUh$x6EH]h$x6@EH] h$x6]h$xjZehYEHUjI h$xEHUV h$xEHjh$xEHU jmhlVh$xhlVh$x6@EH]hlVh$x6@]hlVh$x6EH] hlVh$xhlVh$x6@ EH]hlVh$x6@EH]hlVh$x6]dfhj@B  "$*,.H޸ޥʟʖޏޏʖʖ h$x5\h$x6EH] h$xEHh$x6@]jkhYEHUjI h$xUV h$x;@ h$x6] h$x@ jmh$xh$xjh$xUjhhYEHUjI h$xUV8 8h bPj6:4X`X "hA"hA"hA"@"@"@"@""L""@"(Y " "@"k"k" ";""<5""@"""@"@"( Oh^h`BP & FHJnpz|LN vxz46PRTVX`b|~管痑zmjI h$xEHUVjVrhYEHUjI h$xEHUV h$xEHjh$xEHUjtohlVhYEHUjI hlVh$xUVjhlVh$xUhlVh$x6EH]hlVh$x6]hlVh$x5\ hlVh$x h$x5\h$x h$x6])~TVXXZZ\hjprjlnprڿڿᤖhlVh$x6@EH]hlVh$x6] jh$xEHh$x@ EHh$x6@ EH] h$x5\h$x6EH] h$x6]h$x h$xEHjh$xEHUj=uhYEHU8 Xv^j VJT D<Pt" ""P"@"DS""DS"@") ") ") ") ":&" ""8"@""@""X ""@" "@B @gdrB VgdheBgdB Oh^h`"$&(*246:<>FHJLNPVXZ^prt~jlnpvxz|~ɻ뱻뭤ɻhlVh$x6@EH] h$x6] jwh$xh$x6EH]h$x6@]h$x jwhlVh$xhlVh$x6@EH]hlVh$x6@]hlVh$x6EH]hlVh$x6] hlVh$xhh$x6EHH*]2 "<>@BVXrtvxvxѾѩєujhYEHUjI h$xUVjWhYEHUjI h$xUVj{hYEHUjI h$xUVj1xhYEHUjI h$xUVjh$xUh$xhlVh$x6EH]hlVh$x6] hlVh$xhlVh$x6@EH],  |~Jܿ蓮sl h$x6]hlVh$x6EH]hlVh$x6@EH]hlVh$x6@] jmhlVh$x5\ jrhlVh$xjhlVhY@ EHUjI hlVh$x@ UVjhlVh$x@ UhlVh$x@ hlVh$x6] hlVh$xhlVh$x5\(JTVprtv  "<>@BD468:PRl¸vkaj hYEHUjI h$xUVjhlVhYEHUjI hlVh$xUVjhlVh$xU hlVh$xjޏhYEHUjI h$xUVjhYEHUjI h$xEHUV h$xEHjh$xEHUjhYEHUjI h$xUVjh$xUh$x#lnprZ\fh"$>@BDbdD ޷ޣޚޏzpi h$x;@jǩhYEHUjI h$xUVjhYEHUjI h$xUVh$x6@]jhYEHUjhYEHUjI h$xUV jhh$xjhYEHUjI h$xUVh$xjh$xUjThYEHUjI h$xUV&f"FHnD Z    " V  t$"j "hA"/"DS""@" "@""Y"@""@"(Y """"@" " "@"") "TA O hh^h`BgdBB @gdrD F J T V       2 4 6 p r ~     . 0 l n       " $ > @ B D  f h j l  uhlVh$x6@]hlVh$x6] hlVh$xjhYEHUjI h$xUVjhYEHUjI h$xUV jwh$xh$x6EH] h$x6]juhYEHUjI h$xUVjh$xUh$x h$xEH h$x@*      VX8:<>RTnpȼwmbjI h$xUVjJhYEHUjI h$xUV h$xEHjhhYEHUjI h$xUVjh$xUh$x6EH] h$x6]h$xhlVh$x6EH]hlVh$x6]jdhlVhYEHUjI hlVh$xUV hlVh$xjhlVh$xU#R8L  D H   "/"/"/"/" ""TA "" "")")")"""|"|"","/n  Bgd+ P & F hh^hBP & F hh^h`B VgdheB gdheB gdheprt8bdfhLNhjlntvz~  $&*ɿ鸲駝閸鏲www jGhlVh$x hlVh$x jwh$x jGh$x jrh$xj#hYEHUjI h$xUV h$xEH h$x6]jhYEHUjI h$xUVjhYEHUjI h$xUVh$xjh$xUjhYEHU,*,PRT $*, öwmcUcjh+h$x5U\h+h$x5\jMhYEHUjI h$xUVh$xjh$xUh$x56;@\] h$x5\h$x56\] jrhlVh$xjhlVhYEHUjI hlVh$xUVjhlVh$xUhlVh$xEH jGhlVh$x hlVh$xhlVh$x6]!     " < > @ B D F H ^ ` z | ~      X!n!p!","."""ƵƛƆvlb[bN[bN[bh$x56;@\] h$x5\h$x56\]jhYEHUj h+hY5EHU\!jI h+h$x5UV\h$xjhYEHUjnh+hY5EHU\!jI h+h$x5UV\h+h$x5\jh+h$x5U\jBh+hY5EHU\!jI h+h$x5UV\ X!""#$'()V+*////0444:55L67H7778>9<=>"|")")")"PM"/"T""/"/"(Y "/"d" "/"5"5"5"/"5"̔"̔"/""/"/"/"/Bc n""####"'$'&'(((((())))T+V+X+.(/*/,///////////000d3f3v3x3334444ڻګګڥښڊګګګڃ jh-O h-OEHjhYEHUjhI h-OUV h-O@ h-OEHjhYEHUjZI h-OUVjh-OUhPh-OCJaJh-O h-O0Jd hlV5\ h$x5\h$x56;@\]044444444:5<5>5@55556L6N6P6R6V6X6\6^67 77J7N7777777888<9>9<<"<=====>>>>>>>>>>>>>>>>>>>>>>߾hhemHnHujh-OUh# jh# U h-O6] h-O0JdhP h-OEH h-O@ h-O@ h-O h-O5\E>>>>>>>>>>??(?*?,?.?"/")K$a$M$a$>>>>>??????$?&?(?,?.? hlV5\hhemHnHujh-OUhPh-OhP0JQmHnHu h-O0JQjh-O0JQU21P<0/ =!"#$%rr*2wDd $ hb  c $A? ?3"`?2blCp| D`!blCp| @|cxڕToAfh kQ=T$$&İbcCïࢫ@ipR1yz7iMxh?x0rlt{2, TB;77of&)mb3N@!vsOQ59E͉itKpQG»ToMdHU&z\\ǙiBc•W+W,k5f*eu_^j|4݊Ab;hnM+c]-IHڭɅ .7q#p;9P-u}G=mC{?_ 3@`2W2Nr>?ߛ=1ᾱlp8E/f7 ,arh oCwpdo5p"8[,\ΊWⲫYzvz]Hfd/L3ZDz2eGDdK5ʞzf9RkUxUj~,!_{KfqXC*_aY~Kt*Qi*ňkb!}}vSQyu9꽙hTЭw'sG&\bܐfgsʸ|SQ!:͠t J&!ȓr)hCkw$^Z&6i19y`01X/JPۅ`^K BBiB}lS=WemV3vlG45NX_4pp/" FBvtT߬t>YfWYqpUc&gbAՔ<UpE>-{ɩjakSDd J  C A? "2K Y.Y㓄Zs `!K Y.Y㓄ZslRxeRMKQ=0&DP@֎FTZN 5G"!hk?"R6{N ==aCp mdwGzlMz8JpQ3dB; K(A% 14]ӍD]U&P,>Zg<()~IQxd.TVJm4ZE=flu,Lya\V7 \Wy / *9ZZ1V=۾0ƥEWqV y훎evD!^4: C[(ʶZsspT`m{ JVQ Hzd ]=miJQӢl/j4C8xU6;MK4+nl_d6MQDo~rfDd  hb  c $A? ?3"`?2ԾɆ-q=[I`!ԾɆ-q=[N@|RxڝSOQ۶fۂ&&h`ⱱK@ҔqMԴK5F9X+\4Ɓx/^.'m$fgy73;!H9 I8cxuIvy/.S<@օ!tE0%.$2Opa/*|Ūwt  vtLҋTR*+x[kDQ[%^0  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~Root Entry F&R@"Data WordDocument4xObjectPoolp%R&R_1233261654Fp%Rp%ROle CompObjiObjInfo  #$%&'()*+,-./012345679:;<=>?BEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdeghijklmpstuvwxyz{|~ FMathType 5.0 Equation MathType EFEquation.DSMT49q { _.  @&N` & MathTypeP "Times New Roman؟ww wt fo-OlePres000 Equation Native f_1233261655 Fp%Rp%ROle 2 / ]>Times New Roman؟ww wt fo- 2 ()p.Times New Roman؟ww wt fo- 2 QAKRv<Symbolv -؟ww wt fo- 2 ab"Symbolu *؟ww wt fo- 2 ----Symbolv .؟ww wt fo- 2 =+Q&VMathTypeUUJDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Q==A(aK "- ++bR "- ) "-/ & "Systemot fo !@-KJ( DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Q==A(aK "- ++bR "- ) "-/ FMathType 5.0 Equation MathTyCompObj iObjInfo!OlePres000 "LEquation Native 8pe EFEquation.DSMT49q${$ .  @ &N  & MathTypeP "Times New Roman؟ww ww fd- 2 2y@Times New Roman؟ww ww fd-2 where , ~>`Times New Roman؟ww ww fd-62 , , 0, ( ) 1, and 1 0.```Fr`N`r3BTimes New Roman؟ww ww fd- 2 Ay` Symbolt 3؟ww ww fd-2 >+=-<&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  where A, , , >>0, ( ++ ) == 1, and "-1<< `" 0 2 . & "Systemdw fd !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  where A, , , >>0, ( ++ ) == 1, and "-1<< `" 0 2 ._1233261656Fp%Rp%ROle @CompObjAiObjInfoC FMathType 5.0 Equation MathType EFEquation.DSMT49q&@  .  @&N` & MathType -9 -OlePres000DhEquation Native f_1233261659 "Fp%Rp%ROle nM  sK Times New Roman؟ww wf- 2 d(/)itTimes New Roman؟ww wf- 2 d(/)`Times New Roman؟ww wf-2 |constant.$ll8Times New Roman؟ww wf- 2 // "Times New Roman؟ww wf- 2 KR= "Times New Roman؟ww wf- 2 KRs Times New Roman؟ww wf- 2 QQt<Times New Roman؟ww wf- 2 KR`Times New Roman؟ww wf- 2 QRTimes New Roman؟ww wf- 2 KRQQ=`Symbolv ؟ww wf- 2 sR`Symbolr ؟ww wf- 2 ==a&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   s== d(K/R)K/R d(Q K /Q R )Q K /Q R |Q==constant. & -"Systemf !@-K( DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   s== d(K/R)K/R d(Q K /Q R )Q K /Q R |Q==constant. FMathType 5.0 Equation MathTyCompObjoiObjInfoqOlePres000rEquation Native }pe EFEquation.DSMT49qOl 6{.1  &N@ & MathTypeP `.Times New Roman- 2 Qc"Symbol- 2 -{&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   Q  "- & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   Q  "- FMathType 5.0 Equation MathType EFEquation.DSMT49q{N '.  @&N & MathTypeP  _1233261660Fp%Rp%ROle CompObj iObjInfoOlePres000!vEquation Native [_1233261661($Fp%Rp%ROle "Times New Roman؟ww w fV- 2 GTimes New Roman؟ww w fV-2  with (`llrTimes New Roman؟ww w fV- 2 )1.Times New Roman؟ww w fV- 2 QKRSymbolu ؟ww w fV- 2 =+= &KMathTypeUU?DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Q==K  R   with (++)==1T & "SystemV fV !@-K?( DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Q==K  R   with (++)==1 FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObj#&iObjInfoOlePres000%'Equation Native { .  @@&N & MathTypeP "Times New Roman؟ww w f-2 /I ]>ZTimes New Roman؟ww w f- 2  (DTimes New Roman؟ww w f-2 +) with ()1'X`llr.Times New Roman؟ww w f- 2 QAKR4A Symbol Ҍ؟ww w f- 2 by"Symbol ٌ؟ww w f- 2 ---  Symbol Ӑ؟ww w f- 2 =+=َ&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Q == A(K "- +bR "- ) "-/  with (++)==1 & "System f !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Q == A(K "- +bR "- ) "-/  with (++)==1_1233261662*Fp%Rp%ROle CompObj),iObjInfo FMathType 5.0 Equation MathType EFEquation.DSMT49qO .  &N & MathType F"Times New Roman؟ww w f;- 2 1/wX"Times New Roman؟wwOlePres000+-0Equation Native _1233261663F0Fp%Rp%ROle  w f;- 2 /T "Times New Roman؟ww w f;-2 118ADTimes New Roman؟ww w f;- 2 ()().Times New Roman؟ww w f;-2 QaKLbENJb'\"Symbol ؟ww w f;- 2 -/T 2 ----Symbol ؟ww w f;- 2 N 2 =+  2 N&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Q==a(K  L 1"- ) "- ++b(E  N 1"- ) "- [] "-1/ & "System; f; !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Q==a(K  L 1"- ) "- ++b(E  N 1"- ) "- [] "-1/ FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObj/2iObjInfoOlePres00013^Equation Native  WF6 .  &N & MathType-n nTimes New Roman؟ww wf- 2 1ZTimes New Roman؟ww wf- 2 1ZTimes New Roman؟ww wf- 2 Z`Symbol ؟ww wf- 2 sZ`Symbol W؟ww wf- 2 =Z 2 +Z&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  s== 11++ & -"Systemf !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  s== 11++_12332616646Fp%Rp%ROle CompObj58iObjInfo      !"$%&'()*-0123456789:;<=>?@ABCDEFGHIJKMNOPQTWXYZ[\]^_`abcdefghijklmnopqrtuvwx{~ FMathType 5.0 Equation MathType EFEquation.DSMT49q[ .   &N & MathType-+ 6 "Times New Roman؟wwOlePres00079Equation Native #_12332616654@<Fp%Rp%ROle + w f- 2 12"Times New Roman؟ww w f- 2 02 "Times New Roman؟ww w f- 2 22`Times New Roman؟ww w f- 2 L2Times New Roman؟ww w f- 2 12Times New Roman؟ww w f-2 (1)(1)R``"Times New Roman؟ww w f- 2 T2"Times New Roman؟ww w f- 2 T2mTimes New Roman؟ww w f- 2 UUU `Times New Roman؟ww w f- 2 WU`Symbol ؟ww w f-2 =++++ F 2 ++++ &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==U 0 ++ U 1 1++++ U 2 (1++) 2 ++L++ U T (1++) T & -"System f !@-K( DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==U 0 ++ U 1 1++++     )" !#'$%&*(/?+,-.0312548679<:;=>@CWABDGEFIHMJKLNROPQTSYUVXnZ^[\]_`adbceifghkjplmoqyrstuvwx{z~|} U 2 (1++) 2 ++L++ U T (1++) T FMathType 5.0 Equation MathType EFEquation.DSMT49q  uCompObj;>,iObjInfo.OlePres000=?/Equation Native Ls.  @ &N`  & MathType t` "Times New Roman؟ww w f- 2 yv"Times New Roman؟ww w f- 2 0y Times New Roman؟ww w f- 2 ()ed"Times New Roman؟ww w f- 2 tyt "Times New Roman؟ww w f- 2 ty("Times New Roman؟ww w f- 2 ty"Times New Roman؟ww w f- 2 ty Times New Roman؟ww w f- 2 WUCtO+"Symbol q؟ww w f- 2 =t 2 - 2 = Symbol1 ؟ww w f- 2 =Symbol r؟ww w f- 2 &cMathTypeUUWDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==1U(C t )e "-tt==0t==" +" dt & "System f !@-KW( DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==1U(C t )e "-tt==0t==" +" dt FMathType 5.0 Equation MathType EFEquation.DSMT49q  .   &N_1233261666BFp%Rp%ROle RCompObjADSiObjInfoUOlePres000CEV&Equation Native su_1233261667:RHFp%Rp%ROle y` ` & MathType "Times New Roman؟ww wn f- 2 0y>"Times New Roman؟ww wn f- 2 0yTimes New Roman؟ww wn f- 2 dy"Times New Roman؟ww wn f- 2 ty"Times New Roman؟ww wn f- 2 ty@Times New Roman؟ww wn f- 2 SSRFb"Symbolf G؟ww wn f- 2 ty* 2 ty> 2 ty{ Symbol U؟ww wn f- 2 tyA"Symbolf H؟ww wn f- 2 =y>. 2 =ySymbol V؟ww wn f- 2 =-*Symbolf I؟ww wn f- 2 -&eMathTypeUUYDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  S t ==S 0 "-qR t dt t==0t==t +" & "Systemn f !-KY DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  S t ==S 0 "-qR t dt t==0t==t +" FMathType 5.0 Equation MathType EFEquation.DSMT49qp .1   `&N  & MathType` wCompObjGJziObjInfo|OlePres000IK}Equation Native 0&"Times New Roman- 2 t"Times New Roman- 2 t`@Times New Roman- 2 SRW"Symbol- 2 `Symbol- 2 =-,& MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S  Ԇ"  t =="-R t & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S  Ԇ"  t =="-R t FMathType 5.0 Equation MathType EFEquation.DSMT49q .1   &_1233261668NFp%Rp%ROle CompObjMPiObjInfoOlePres000OQEquation Native 0_1233261670LXTFp%Rp%ROle Nf & MathType0 `Times New Roman- 2 d/dw&"Times New Roman- 2 t`@Times New Roman- 2 SStW"Symbol- 2 `Symbol- 2 =&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S  Ԇ"  t ==dS/dt & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S  Ԇ"  t ==dS/dt FMathType 5.0 Equation MathType EFEquation.DSMT49q. CompObjSViObjInfoOlePres000UWVEquation Native .1  &Nf & MathType0 '"Times New Roman- 2 .w."Times New Roman- 2 t`HTimes New Roman- 2 S`:Symbol- 2 -&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  "- S .  t & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  "- S .  t FMathType 5.0 Equation MathType EFEquation.DSMT49qEpn .1   &_1233261673ZFp%Rp%ROle CompObjY\iObjInfoOlePres000[]Equation Native D_1233261674.`Fp%Rp%ROle N@ & MathType` "Times New Roman- 2 .S"Times New Roman- 2 ttt`FTimes New Roman- 2 KQC`Symbol- 2 =-&4MathTypeUU(DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K t. ==Q t "-C t & "System-p(~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K t. ==Q t "-C t FMathType 5.0 Equation MathType EFEquation.DSMT49q p bCompObj_biObjInfoOlePres000acEquation Native q.    &N  & MathType` "Times New Roman؟ww wf- 2 .y`Times New Roman؟ww wf- 2 (,)qt"Times New Roman؟ww wf- 2 t ""Times New Roman؟ww wf- 2 ttt:`FTimes New Roman؟ww wf-2 KQKRC`?Symbolm ؟ww wf- 2 =-&aMathTypeUUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K .  t ==Q(K t ,R t )"-C t & "Systemf !-KU DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K .  t ==Q(K t ,R t )"-C t FMathType 5.0 Equation MathType EFEquation.DSMT49q  u.  @ &N`  & MathType t` _1233261675fFp%Rp%ROle CompObjehiObjInfoOlePres000giEquation Native s_1233261676dplFp%Rp%ROle       "%&'()*+,-./01235678;>?@ABCDEFGHIJKLMNPQRSTWZ[\]^_`abcdefghiklmnqtuvwxyz{|}~"Times New Roman؟ww wf- 2 yv"Times New Roman؟ww wf- 2 0 Times New Roman؟ww wf- 2 ()ed"Times New Roman؟ww wf- 2 tt "Times New Roman؟ww wf- 2 t("Times New Roman؟ww wf- 2 t"Times New Roman؟ww wf- 2 t Times New Roman؟ww wf- 2 WUCtO+"Symbol ;؟ww wf- 2 =t 2 - 2 = Symboln ؟ww wf- 2 =Symbol <؟ww wf- 2 &cMathTypeUUWDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==1U(C tt==0t==" +" )e "-t dt & "Systemf !-KW DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==1U(C tt==0t==" +" )e "-t dt FMathType 5.0 Equation MathType EFEquation.DSMT49qp .1   `&N  & MathType` "Times New Roman- 2 CompObjkn!iObjInfo#OlePres000mo$Equation Native 40.w&"Times New Roman- 2 t"Times New Roman- 2 t`@Times New Roman- 2 SR`Symbol- 2 =-,&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S .  t =="-R t & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S .  t =="-R t FMathType 5.0 Equation MathType EFEquation.DSMT49q pT *.1    &N  & MathType` _1233261677rFp%Rp%ROle 9CompObjqt:iObjInfo<OlePres000su=|Equation Native Ot_1233261678jxFp%Rp%ROle U"Times New Roman- 2 .`Times New Roman- 2 (,)qt"Times New Roman- 2 t""Times New Roman- 2 ttt:`FTimes New Roman-2 KQKRC`?Symbol- 2 =-&aMathTypeUUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K .  t ==Q(K t ,R t )"-C t & "System-pX~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K .  t ==Q(K t ,R t )"-C t FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObjwzViObjInfoXOlePres000y{Y Equation Native j4 .  `&N` & MathTypep "Times New Roman؟ww w f- 2 ,yA"Times New Roman؟ww w f- 2 CttTimes New Roman؟ww w f- 2 UeSymbol ؟ww w f- 2 weSymbol; ؟ww w f- 2 =e& MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  U C,t ==w t & "System f !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  U C,t ==w t_1233261679~Fp%Rp%ROle oCompObj}piObjInfor FMathType 5.0 Equation MathType EFEquation.DSMT49qN .1  `&N` & MathTypep "Times New Roman- 2 ,"Times New Roman-OlePres000sEquation Native 4_1233261680|Fp%Rp%ROle  2 ttRtaFTimes New Roman- 2 PQSymbol- 2 wSymbol- 2 =&$MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P t ==w t Q R,t & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P t ==w t Q R,t FMathType 5.0 Equation MathType EFEquation.DSMT49qph 4.    &N & MathType` "Times New Roman؟ww w f- CompObjiObjInfoOlePres000Equation Native .2 .y`)Times New Roman؟ww w f- 2 q>"Times New Roman؟ww w f- 2 t"Times New Roman؟ww w f- 2 t`FTimes New Roman؟ww w f- 2 PP` Symbol ؟ww w f- 2 =P&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P .  t ==P t & "Systemڊ f !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P .  t ==P t FMathType 5.0 Equation MathType EFEquation.DSMT49q_1233261681Fp%Rp%ROle CompObjiObjInfoOlePres000Equation Native o_1233261682vFp%Rp%ROle   .  @ &N@  & MathTypep o"Times New Roman؟ww w;f- 2 .ys"Times New Roman؟ww w;f- 2 ,`/Times New Roman؟ww w;f- 2 wD"Times New Roman؟ww w;f- 2 t"Times New Roman؟ww w;f- 2 tKttK`Times New Roman؟ww w;f- 2 Q`"Symbolf ؟ww w;f- 2 wwwp`Symbolj q؟ww w;f- 2 =-&_MathTypeUUSDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   w .  t ==w t "-Q K,t w t & "System;f !@-KS( DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   w .  t ==w t "-Q K,t w t FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObjiObjInfoOlePres000Equation Native 4p d.   &N & MathType` "Times New Roman؟ww w f- 2 .y`Times New Roman؟ww w f- 2 /ZLTimes New Roman؟ww w f- 2 Zq>"Times New Roman؟ww w f- 2 tZ"Times New Roman؟ww w f- 2 tZ`FTimes New Roman؟ww w f- 2 PP`Symbol ؟ww w f- 2 =P&$MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P .  t /P t == & "System f !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P .  t /P t == FMathType 5.0 Equation MathType EFEquation.DSMT49q, .    &N_1233261683Fp%Rp%ROle CompObjiObjInfoOlePres000TEquation Native d_1233261684Fp%Rp%ROle  & MathType-@ E"Times New Roman؟ww wf- 2 .+"Times New Roman؟ww wf- 2 ,+@,Times New Roman؟ww wf- 2 +eb"Times New Roman؟ww wf- 2 t+O"Times New Roman؟ww wf- 2 KtS"Times New Roman؟ww wf- 2 tt@ATimes New Roman؟ww wf- 2 QtN@Symbol ؟ww wf- 2 wtlO 2 wt@3Symbol k؟ww wf- 2 +=&TMathTypeUUHDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    w Ԃ.  t w t ++Q K,t == & -"Systemf !@-   !"#$'*+,-./0123456789:;=>?@CFGHIJKLMNOPQRSTUVWXYZ[\^_`abehijklmnopqrstuvwxyz{|}~KH( DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    w Ԃ.  t w t ++Q K,t == FMathType 5.0 Equation MathTyCompObjiObjInfo OlePres000 BEquation Native  9pe EFEquation.DSMT49q .    &N & MathType-@ "Times New Roman؟ww w[f- 2 .Z@Times New Roman؟ww w[f- 2 Z "Times New Roman؟ww w[f- 2 tZ "Times New Roman؟ww w[f- 2 tZJdTimes New Roman؟ww w[f- 2 PZldTimes New Roman؟ww w[f- 2 PZ@Symbol^ ؟ww w[f- 2 =Z&)MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    P t. P t == & -"System[f !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    P t. P t == FMathType 5.0 Equation MathType EFEquation.DSMT49q_1233261685Fp%Rp%ROle %CompObj&iObjInfo(ph 4.    &N & MathType` '"Times New Roman؟ww w f- 2 .y`)Times New Roman؟ww w f- 2 q>"Times New Roman؟ww w f-OlePres000)Equation Native <._1233261686Fp%Rp%ROle A 2 t"Times New Roman؟ww w f- 2 t`FTimes New Roman؟ww w f- 2 PP` Symbol. ؟ww w f- 2 =P&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P .  t ==P t & "System f !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P .  t ==P t FMathType 5.0 Equation MathType EFEquation.DSMT49qV  .  `` &N  & MathType` 7CompObjBiObjInfoDOlePres000EEquation Native ]Y"Times New Roman؟ww w f- 2 *yY"Times New Roman؟ww w f- 2 "Times New Roman؟ww w f- 2 0STimes New Roman؟ww w f- 2 e"Times New Roman؟ww w f- 2 t)"Times New Roman؟ww w f- 2 ttFTimes New Roman؟ww w f- 2 PPP  "Symbol ؟ww w f- 2 -tSymbol. ؟ww w f- 2 ==&IMathTypeUU=DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P t* ==P t e "-t ==P 0s & "System f !-K= DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P t* ==P t e "-t ==P 0 FMathType 5.0 Equation MathType EFEquation.DSMT49q! _1233261687Fp%Rp%ROle cCompObjdiObjInfofOlePres000gEquation Native F_1233261688Fp%Rp%ROle .  `&N  & MathType-@r   "Times New Roman؟ww w f- 2 .yMATimes New Roman؟ww w f- 2 y$"Times New Roman؟ww w f- 2 KyMTimes New Roman؟ww w f- 2 QyNLTimes New Roman؟ww w f- 2 CylLTimes New Roman؟ww w f- 2 CylxSymbolh V؟ww w f- 2 hyM9SymbolY Č؟ww w f- 2 -y@ 2 =y&6MathTypeUU*DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    C . C== Q K "-h & -"System f !-K* DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    C . C== Q K "-h FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObjiObjInfoOlePres000B Equation Native  p    .  ` &N B & MathType0-@r@r @ r 7"Times New Roman؟ww w^ f- 2 .Z!"Times New Roman؟ww w^ f- 2 .Z "Times New Roman؟ww w^ f- 2 .ZVTimes New Roman؟ww w^ f- 2 0ZUTimes New Roman؟ww w^ f- 2 Z@Times New Roman؟ww w^ f- 2 0ZUTimes New Roman؟ww w^ f- 2 Z* Times New Roman؟ww w^ f- 2 0ZOTimes New Roman؟ww w^ f- 2 Z"Times New Roman؟ww w^ f- 2 KZ"Times New Roman؟ww w^ f- 2 KZ "Times New Roman؟ww w^ f- 2 KZdLTimes New Roman؟ww w^ f- 2 CZVTimes New Roman؟ww w^ f- 2 QZLTimes New Roman؟ww w^ f- 2 CZNLTimes New Roman؟ww w^ f- 2 CZ@Times New Roman؟ww w^ f- 2 QZl LTimes New Roman؟ww w^ f- 2 CZ8 LTimes New Roman؟ww w^ f- 2 CZ* Times New Roman؟ww w^ f- 2 QZVLTimes New Roman؟ww w^ f- 2 CZVSymbol| O؟ww w^ f- 2 >>:!@ 2 ==:!*  2 <<:&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    C . C>>0!Q K >>  C . C==0!Q K ==  C . C<<0!Q K << & -"System^ f !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    C . C>>0!Q K >>  C . C==0!Q K ==  C . C<<0!Q K << FMathType 5.0 Equation MathType EFEquation.DSMT49q_1233261689Fp%Rp%ROle CompObjiObjInfoOlePres000Equation Native @_1233261690^~Fp%Rp%ROle ] {p .1  @&N@ & MathType` Times New Roman- 2 (,)&"Times New Roman- 2 tttNTimes New Roman- 2 RS4Symbol- 2 G=G&&0MathTypeUU$DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  G t ==G(R t ,S t ) & "System-p$~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  G t ==G(R t ,S t ) FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObjiObjInfoOlePres000Equation Native t u.  @ &N  & MathType v"Times New Roman؟ww w fh- 2 0y Times New Roman؟ww w fh- 2 ()ed("Times New Roman؟ww w fh- 2 tt "Times New Roman؟ww w fh- 2 t("Times New Roman؟ww w fh- 2 t"Times New Roman؟ww w fh- 2 t Times New Roman؟ww w fh- 2 WUCtOtm "Symbol7 ؟ww w fh- 2 r+"Symbol ؟ww w fh- 2 =t 2 - 2 = Symbol7 ؟ww w fh- 2 =Symbol ؟ww w fh- 2 &dMathTypeUUXDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==1U(C tt==0t==" +" )e "-rt dt & "Systemh fh !@-KX( DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_     !"#$%&'()*+,-./012356789:=@ABCDEFGHIJKLMOPQRUXYZ[\]^_`abcdefhijknqrstuvwxyz{|}~A  W==1U(C tt==0t==" +" )e "-rt dt FMathType 5.0 Equation MathType EFEquation.DSMT49qp _1233261691Fp%Rp%ROle CompObjiObjInfoOlePres000Equation Native 0_1233261692Fp%Rp%ROle .1   `&N  & MathType` "Times New Roman- 2 .w&"Times New Roman- 2 t"Times New Roman- 2 t`@Times New Roman- 2 SR`Symbol- 2 =-,&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S .  t =="-R t & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S .  t =="-R t FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObjiObjInfoOlePres000 Equation Native 4Bp \.1   `&N  & MathType` 3"Times New Roman- 2 .`Times New Roman-2 (,)(,)tqt"Times New Roman- 2 t""Times New Roman-2 ttttt:`FTimes New Roman-2 KQKRCRSB`?Symbol- 2 =--G&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K .  t ==Q(K t ,R t )"-C t "-G(R t ,S t ) & "System-pР~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K .  t ==Q(K t ,R t )"-C t "-G(R t ,S t ) FMathType 5.0 Equation MathType EFEquation.DSMT49q={r ~_1233261693Fp%Rp%ROle ;CompObj<iObjInfo>OlePres000?Equation Native N_1233261694Fp%Rp%ROle S.  @&N & MathType` Times New Roman؟ww wf- 2 yA"Times New Roman؟ww wf- 2 C4Times New Roman؟ww wf- 2 U4cSymbol R؟ww wf- 2 =4~&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  U C === & "Systemf !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  U C == FMathType 5.0 Equation MathType EFEquation.DSMT49q { CompObjTiObjInfoVOlePres000WEquation Native g(.  @ &N`  & MathType` Times New Roman؟ww w f- 2  "Times New Roman؟ww w f- 2 RRFTimes New Roman؟ww w f- 2 PQlSymbol ؟ww w f- 2 =-G*% &MathTypeUU DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P==Q R "-G R & "System f !-K  DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P==Q R "-G R FMathType 5.0 Equation MathType EFEquation.DSMT49q_1233261695Fp%Rp%ROle lCompObjmiObjInfooOlePres000p4Equation Native -_1233261696Fp%Rp%ROle ] p  .   &N@ & MathType` "Times New Roman؟ww w fJ- 2 .y`Times New Roman؟ww w fJ- 2 "Times New Roman؟ww w fJ- 2 S`FTimes New Roman؟ww w fJ- 2 PP*`Symbol ؟ww w fJ- 2 =+Gu&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P . ==P++G Sr & "SystemJ fJ !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P . ==P++G S FMathType 5.0 Equation MathTyCompObjiObjInfoOlePres0004Equation Native -pe EFEquation.DSMT49q p  .   &N` & MathType` c"Times New Roman؟ww w7 fQ- 2 .y`4Times New Roman؟ww w7 fQ- 2 p "Times New Roman؟ww w7 fQ- 2 K`Times New Roman؟ww w7 fQ- 2 Q`Symbol ؟ww w7 fQ- 2 =-&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    . =="-Q K u & "SystemQ7 fQ !@-K( DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    . =="-Q K _1233261697Fp%Rp%ROle CompObjiObjInfo FMathType 5.0 Equation MathType EFEquation.DSMT49q< p  .   `&N  & MathType` 3"Times New Roman؟ww w f- OlePres0004Equation Native ._1233261698Fp%Rp%ROle 2 .y`Times New Roman؟ww w f- 2 m"Times New Roman؟ww w f- 2 s`FTimes New Roman؟ww w f- 2 PP*`Symbol 5؟ww w f- 2 =+G&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P . ==P++G s  & "System f !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P . ==P++G s  FMathType 5.0 Equation MathType EFEquation.DSMT49qp  .    &N}\{?N,'HaU1,c0b89OoA5Ϻ"}.d!W|`zɎ Rvnк!Ǫ3%MJ&eq*3y/_+iTƍt^֛bt}a WnYOLe;)`GYM-|=7LiY۝K'kOhsvh;G:e-^U0PG)imHtQI2sC WԂݯ]">ZBBYBUms;s}'n;&0lJnZz K⿉5uPq9`Dd 'hb  c $A? ?3"`?2##^zHuڍA`!#^zHuڍAؐ@@%|xڝTKhSA=3/GiԦ! ]F+%%S}@H YiPDtequ .҅ PtSARUxӏVofܹ; g+! YcivV;<`C }o6ɝdeE0R$q깆ٚ&B?xKAi!eهq5%y ݓu*f1c CYۨ=m.kݸ`1̱+8EVy."ݟx~U DE[|g~CG?vəhzTn]8% UԳbXB9A*G(#J@N:W \H"<5/?Xs5jSKɖ94vBvpԛͫ{ssgbxΈ̫hN#L>o9RXV ?bF#ucJ,IEu'M5q9&;q;Ub$cj: 'j߫4|X }9~a-]aBv[h-ܣNjPp@b2QF)AOYdhq0僯1P5*f$ʰE,&Cgf^RԺi(7/\&xFTQQE5ZҸ0军QDSћA(Tn!wwdRw8~9; MB!BS1S@XnuF(pe8D/ўhX":CJZd 0{qSlN7Ś u.g$TްI%Qx{v#[/]IGG@sP磿ϯde+.,;va)a^ Q=Rq#_;dv S8 vh''Je@ 2'.=OV~0MN>_ F':Xg2yx}o޼a İ941> t?m-E섁 3a6-\PT襣Tk.`v_*SG9{.QPChZjsemb/G?ffZwI_i֝>D3y7qaA2t*W9;F+0Y+f>%>iS!h<w}==zAb-C~G ArreڰjEǫ5o%!7!LZyUu[҈/:n#B??o|Ӹl[oN>fX h2\^%Z-]g:c0NĨAT؃㱇Yd|. J 2ƿ$_jNDi8nd&4N^ۭV( Q@%9w>'xuC (Dd `b   c $A ? ?3"`?2M)?)2z:en)`!!)?)2z:en #dxڭUKSQιm6QJMPG.ԛԕ*8Wjp S!ł"Uz (!Cs.5rvC |DЁ5 ;%Ĕ-t0uԲРkQ:ꐡ9`X@jXNDZcQ>jQzt!gN-ݣz&)'QE6l%Z68M38mr{0}Wn7w+,iX:6"faűDk*s2_dqѯ9n E2++d Y}Ffk-Va y ?D>ZQ͓;uC$؆ XTƱ8dQ9胪މN+cc*c*EdohV5AH>}&xr?^]3fa؝!lP#_Hh;6qX& N5R݉llO6e''kI=ImM$&RZ.~p6G !+GB9 G@GBw%ÆBToCӅ>"mxTg!3Ba&E#1nQwB'*4M7:0Ӹ;8|Mذ@RLQ87*[Ѽ۶:hWZլ*C;kX`Gٜ`&1fg/g@0VGDd b   c $A ? ?3"`? 2cauՔcQZkX `!cauՔcQZkX @x xڵUOAfvt)`l4&PJ#MªMZbӓ /0Q{1&9Ozdb/'MkQy`ҲplA<6j1KbRvu𪝃I]$os‹04\!9yftXVL%s[*ix \H'/I?bM 5L9ߴ]ͤKx[5{zLCdMpQsԽ63};r 8o1Rm&ϣa`aM_]eqH&>I|hou>a@qhQy4> kWloK^Vk BZ&ݏQjXb#KWPѲ8,3 э)OK  w'I ']! \0'H"r7JpZ^@Ha2=CG(8h~WU1R$;Io#*y_B'p n\8amrF$Bjƾxм$c/ kbDd b   c $A ? ?3"`? 2^`c3/s$`!^`c3/ Xy(+xڭoA춰bD)0Gn‚U>55x@F/Mo^[M<\ov+4a2oyo~o7lP(!Dh5rš;B{v.Jg'.7eʼH+\k72͛*wϽ8?ʥc7̏wہG(x2ŲZ,@ZW/sm@K|_:&y|}b c׸7c[B#Hc'W `ո~ ʠ(鋯0sq ᑔfyZXsIKO>Ʀa,,q{'轍><akӷdȿx_~ ^6"wGb~r>^fHcJ2s`J"JzZZPSj.{ޮW&󵒪ilSǽSӧP>Agb%;ʻV\z˘Y~=rqd a.10*l*8<hHL@Dls}/fe$w3=fiw q_skxGqp.yU1(Xhx,4Z]~LC;|`EX&kZDd [J  C A ? " 2N)$71~y{*(`!")$71~y{` xRxڕSkA۬i`7k(6 )=nV]j i$Y)9)zM/Ŀc7OBo *{\̮Z-:7|f0  `Ct&)8IKWKFs 1:悠K#Q[/Ӓ\J(ZWDn?9[LqY²ϫ4R9l X;ŏטYնFʻCu4bS}F㟹RC0Ssשmo/)k^,t^o#uUfP4VhdٰUEn^Ḙ6czo7o^\v<VUI=^7})]k~=VkHB,M6DApy82`xr3U /?+J5TH*aϗm\VҢgi#Jxwy?nRDd tJ  C A? " 2a6en*f4= +`!56en*f4  @CxڅSMkQ=NӚt`f_1`M\:Mݤ'8QISZ۸(D. W ؚ}w}gλ>B0>؇S2qk-=YR5e$C Zpڄu9NiMLic˚%X yA?/- IhwA~=uDMbT!Ul ŘsiYSN:&i?jOJ`R)tJ=/cSśAb,ް_ >]p$"ۻBml8fWk6[nڏʽVP8Z8FW'S=뷃 a8wXUn={MU󖋨݈FMAUVfTkH{V>_.5+j>2IA,CLr<l9`CUͷ>$lrh( xjek[D U\q"K3zuߏ*xAtO~fDd J  C A? " 2'QP@v .`!QP@v.@C@CxڍAkQ3QaWPBKc&PB/=uR/7v n * x~OKy/kҼw{UFc 5b2qajO6KV)F;=w~_}7@&yƈv˽Wpw{q_F̦{3){[tmP2Q]gYhoPJJ̕FyX|:ztuG. gͥi&r\zm_f7׫ʺ]-k׿Xdp㊲LGQ5\swÑ^bWVj^̒XI$Q:ٌRy0^^ғ@Oо<;|$gx欌\vU ̝pXד6&uֳ$BcJ7]gDd lb  c $A? ?3"`?2'L E bw3`!'L E bw XhSxڥTMkA~f&&IZZm@iKEMVW $&+!J^ң7?!A zlĈXswI@lr 949c.b9(f\\S.TEOcA& ?Y;FlY)J*ɠpϓ8 a{6 %͎-#Q&~^ 8HEreV+OR\p1nK}Dd b  c $A? ?3"`?2?вÌ%7`!?вÌ @x xڵUkAf66%?4U𤁦ibrLcj iZr {7o^m\ORMVKKf{{{;3 @5N3fIW*C:NZ6^sa鉳-.t"17i6H~Nc |%6ʉhJ5Y%#CB:U6%w+F:(TRztVƕߵ\65w-|}_rd#Y#\p.336c#6A[a /ZLşÀ^>ef!X;x+5aZ㣲7a쇱{[6$5DsˣxXj^[f+V^e򥇛޿kXb#]hd-KkA^Ol]АݬCnjN;*ЈQmki5ͪU<`vFo4t$ҳLf$OOs3Dꦒide㱇ssieN,ãg0R Mnוdh|~F/$Irb*F^444#!ԟMFt h0NՂ= F9eQm5yZqzRLOQmU3 JqYe[khy_Lh}IX]"5?@WX-X(q&1ߺq-W,J =xE$oDd [J  C A? "2Ia.ZGsq%:`!a.ZGsq˦` xRxڕSjQΙi`&MHŎ(&E@IIq9u@H2RRZp7EAp3NpVh~r1A-i~8r#5i%K&xUnʹ3ͲHos%G*+RV'kw_zKVULـq^&޸x7[t`[ȔqhoGC F'Ӱ< Gțy:x9CeDFpg>mضnmߍ' ۪*1Ǜ^<њ~ͯuI呄H4KU#Ɖ'3lqYXS:e]Rμ"bI-̹IWa:6ݙCR?QC)Dd lJ  C A? "2rZU`[g=`!_rZU`[T Xh-xڕSkQ޼$I`7U1IQ"&BHIɪFɩO=ԃ wxO"{M$o{3󾝙7#$EB Rx<֚X՚k$P0AQϑeFWC.`a&k>0~m'"( ͲHdXac-fvzyi{SlLp=ܔsgk~g{qxR6wNY%uf& }H?#L4⏔/"v]_Mw0[GlL[aPbU9:#jҠ[dכ*N b6WA&uMzGRD[rb|gZ[v yGNl19o,<62"FEr@KX6gwVǥW ^޻ L#0%Pknf`%zSeD_ ~8!Jd l ن%+2a ]VcǸam`Dy9jz[4qJsRJ:&Dd |b  c $A? ?3"`?2pי~60^6LA`!Dי~60^6` 0xڝSkQm?+IU=ڂIn7^RI]u@Ҕd˒EC z(C/RГ v=cٕ-73ߛyӀ>PYQHDQe:͈N:℁9D0y,l83߯<`+2- NC@s(>(Vfu/hMyve?]3fؚ3eM:Iץ>b?&H PL;tCDŽƲ{PR8HrT{[&T^rW`Hs`q\'qҎ׉T*r 0{i5nqn*=:^ì&NMdE4=GS+AqŪqݱ*Οc|ʰ `adTR謅xk EBYN \ dmyz[>{$Ϲ4yR@AdV.F_|OGO&&~ ^Z0e͙إrk%vDd $|J  C A? "2S9Z H]6/*D`!'9Z H]6`E 0xڍ1oPwMP*@P*` U:FM]RI*F "%M25l] XG``7$̽W'iJ`پw΄ FJD0q=,j5HT7CzJԃ["gHiY1DeQ׵vlkʄR;Ӣ~l{҆a͠3tg*dP<Ψ0אּ`/T/?H !B0M jBJG-UOgmBÌK ,{/5bd8j9RM0"5v4Dd b  c $A? ?3"`?2؉51I@V?!'QJ`!؉51I@V?!' @F kxڭTMlAfEc4JIHL<(*4)V1Pl8Yo護jz/xЋڮ'[nvvμy S&P,;mޘ44YGG:c^&- |wKCh7k:PL|\ZxP'v;oTE ܨVklS ~\mț*dhźwxu c}]_#g] 6a$Va y`#m|i6s\4l1I\rZV5'?h A4TH>B" ٨$yEbw sLaPx2_o|QѳŻz>e\Kǧ$%.'bpDRd$xvO{R%5_ IȮf VQhl!YHY{4F:0NbDT&JL@x+L$w&9* R'Iɠrdz(ߋnD0DKLgcpV,Y׺S oˮD)k6 Gs3xAH3CDd b  c $A? ?3"`?2ΊTlRF%hiM`!aΊTlRF%h L /xڭTkAۤ&Rۢ & xM=$UIJIAz<P< ^=ų?֓`|Q|;{}fcE3QRt(]M cSN822-m1|-;&C Uz{*>˄c pAj m.9bQ]û'_:拏ΣU#"f'Z2)]mR[S- +;MP?|W.h#&ꀙ?@ ghX/ѻ+^qD7㗏 ðV{CCmO@ u":0G6v$cC k jyy E Q ԥښU/T۵+cW9m$&0D4ۼժ9-Hb|6Җōe|/z eAq#^ޛ//z=欌X\92b$iK-H +4mݳ循5x=%fԜZtEXO#"30~}c)\v ug|Hb4WDd 4b  c $A? ?3"`?2xBPpOxQ`!xBPpOx,   pxڵTkAffI7?n zhj @iP&cHf%KM?=GOxēPzQ0&ijuf>޼ޛ~@5ȡk!ƛͦD_ t/hJgd&+/ E[&?YRqُMtL~4J/>t17 RŬY_Vx>]Q绶05$Cu~..6GR\8d.p9L>zuմr8!=emcyCvo䘵dL$Gn2_-Oueezߵ8u;u+ Z<ϼ{>77TO2tryX+-W͙u[h'J\DgKeӲh,ݘ3fpR>}ѐ;mBPJN=R%IIHaՁq$PG&Q@I.k#H~%N5Azd%W-rfN{ W#,oe(Hp{-_QֺOگ7s9 c]wrC7s31'8YS@+ܭt.BdARq*"I4+?=o1? ZnvǸlX[3u޹>s{aR=.e M;+感rr:Gs gvXi:W&ktFk嘦c̑.˞3Q6n6h y@y6ľ4"X'.e*ݨo@]EIx$:)9Bmh]s홗-gO2ͯ'`"bkVI$]sǮӢ6}>P)O\1-:'&{!&8\-7ki ~?VjtAnT[wa,[1<65Kmn^`f2̳|40sj˕VjqM*7"ɭ'"gU)D+͚n?eU ΗT뱣dφT9[.e˙Ҋٳ=giB,eK9SPG` u65ZLJƦv/IF s`JwurDŽMVrb&wtwUt*ܞ 〽/)(tZ^GWyhHv/2n,Dd  b ! c $A? ?3"`?2%i+]ݰHם^`!%i+]ݰHםź`hn 8{xڥTKkSAfMrmP (6 jBjZUz5$ NwV.]T \ng#C 9;ܙ 111-`4w%O/G,oLN"H&sSv;rb.$?*&# C:I:&cq#bwt5yt sܒw`ح_nԀPGe㧕6^xq:%ioy&cLaPx[:IBuEʭtQ(_7Z)JCJqY5b!+7kF03^fV/^v0f]=v \KqhzywX0ݟNsI(^XUb`9y+uoPG:k!z F!JQ5ԏm'AE@q1sDD" =Q(8N}f擵:3~靯vv^A!tϖRw0x>>4{xsb#kDd OhJ # C A!? "2MIP~&=y_)Mf`!!IP~&=y_p@* |xڅSkAƦi`7qEkҖZc $Mnu@HRrj5\shR%| ,YJ-U|ޝl(I֙;߶X~/6ȫ}ChBZj-#!YX 9y#򰰘:Uh9 cZO6)4\xTzϾ,{P]*eo25U?"]O01|= Õ}+쓻yFժRX&;`93WY~;U <3!ʂfV)V?[SW\$Iw6s9WOgsEsd>:hG#s=&'#f\0# 1/1fM0H%F1CʈP.ѓ`SYWK\6ϣdzjSS6eKxȩWSBX4Q.И?8wRފ,S6̾A ֡ {zq[7o*t-YQI Vg D7nw9O֠.^V)24c;MgX 5sYVO蓮uX^;sMxqѠoJ.8KA;<,)5Ѿv^o/uWOfp7fup2CeDFVt{>mضn-k߉'{ ۪*1[~<њ][~ͯuI呄H4KU#+OfTxieYXW:e]Rμ"bI-,IWa66T! GQ_쨡~:-[Dd | J & C A$? "!2P OE~aэo`!P OE~aэθ` |_xڕTkQI4URӀ-M!BJxFɩko^SɓBOIdoyVҨc߾ߛc OBH^ϖDqږ\%D}䧐('yU=yVAcGyxي~bm1GMuS)ܳ7h/ZZ7ZUy(ߨW6GXϙ54QALlh?NsY<hz+p|<%#FM7O(rel o  q]#<3k|"_i蜈+{n}t߀B~Q.%H$}ݦ̹xEMKBx`nuCVAU-xs 'x֬Me+͚ r\sKzSVʺbu%FK7e-V- },,##ܑ`|7/3|SF:Q'࢔Im#u b_cag96 :Kr[ #,LP'Rz+҇^R?sϽ1bv '\hL~mDd hb ' c $A%? ?3"`?"2Pk'WaNXY" x,Js`!$k'WaNXY" xr@]|xڕRKAlj &BQЃ ǀ&%LBۭ]@bJ hJ^PDճAR?RdA曙7{{0hG@  zDwûQ%Ē8120@c &ߜ0.hh@m(u^zeD.&?*P9颋nݺrэwC$s{7=( ? "QHL&]݌yXcE*|- vpf{P< JT Cz'#/U(9DٲU[&>;јcvu2|xxV=ͷ'u_yV.9tb92eHcKHy ZǾS'"ys[A.Pسu"(*8Vyоk}dd:Ϥ bJF GQJ$E8v/FEtr0]|܀[rrQ"^PbTrw,|6#ׂ/q^1U&VCH&1]47/߷*tC% !GQv$qL(bIUL29^1!2%TB~Qv7?H^Dn3& wzT)W V[7Ӎ[]OJ/?lVDfiCp~1,ZkDr%z nֽKHZ^֍Vܞ>Yby/eJYWV$,&)R蔩 Wwk|`!ce t0-fsА7>  ^ 1xڕSKkQLL"ɤi. IAWN`$&e2QISn{?tQ rc\ܛI`{ι! (Ap@,6gC+Q"|l,4Mb0p1V}/hB[xVQJj2JXk~+'L x޴;Ɗ3:?D0~Fod"kbKpŮ^B^rp#2 y#b>>˘U|c%y#1ol56|HvC4<:@^96~:ST+*WuX:l]誟xlk]B vv@hf>?͛E'V5.'|"f;`J( 1;s%g OH["`Usռ#>3eTi!FȂj4dS:0Ml5xATD2űU"34(okL%EdzČFkcJɹ %Nn$@1m6Wo^Ӕ@Dd <CompObjiObjInfoOlePres0004Equation Native . & MathType` '"Times New Roman؟ww w f- 2 .y`.Times New Roman؟ww w f- 2 v"Times New Roman؟ww w f- 2 s`Times New Roman؟ww w f- 2 PP`8Symbol ؟ww w f- 2 =-G4&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P== P . "-G s  & "System f !@-K( DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P== P . "-G s  FMathType 5.0 Equation MathType EFEquation.DSMT49q_1233261699Fp%Rp%ROle CompObjiObjInfoOlePres000Equation Native _1233261700Fp%Rp%ROle t :z.1  &N@F & MathType  "Times New Roman- 2 .`FTimes New Roman- 2 Pz&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P . & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P .CompObjiObjInfoOlePres000\Equation Native   FMathType 5.0 Equation MathType EFEquation.DSMT49qn4 .1  &N`` & MathType g "Times New Roman- 2 ."Times New Roman-     !"#$%&()*+,/23456789:;<=>?@ACDEFILMNOPQRSTUVWXYZ[\^_`adghijklmnopqrstuvwxyz{|}~ 2 0>"Times New Roman- 2 0Times New Roman- 2 d"Times New Roman- 2 t_ "Times New Roman- 2 t"Times New Roman- 2 tt@Times New Roman-2 SSRSRa"Symbol- 2 t, 2 t> 2 ty Symbol- 2 tA"Symbol- 2 =>- 2 =Symbol-2 =-=-,*Symbol- 2 &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  S t ==S 0 "-qR t dt t==0t==t +" ! S .  t =="-R t & "System-pа~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  S t ==S 0 "-qR t dt t==0t==t +" ! S .  t =="-R t FMathType 5.0 Equation MathType EFEquation.DSMT49q3 p, .1   @ &N  & MathType` 3_1233261701Fp%Rp%ROle CompObjiObjInfoOlePres000TEquation Native 'X_1233261702Fp%Rp%ROle -"Times New Roman- 2 .`7Times New Roman- 2 ()w&"Times New Roman- 2 t"Times New Roman- 2 tt.`@Times New Roman- 2 SGSR`Symbol- 2 =-&EMathTypeUU9DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S .  t ==G(S t )"-R t & "System-p<~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S .  t ==G(S t )"-R t FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObj .iObjInfo0OlePres000  12Equation Native B,< p  .   `&N  & MathType` '"Times New Roman؟ww wfq- 2 .y`Times New Roman؟ww wfq- 2 "Times New Roman؟ww wfq- 2 s`FTimes New Roman؟ww wfq- 2 PPPG*R`Symbol ؟ww wfq- 2 =-&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P . ==P"-PG s & "Systemqfq !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P . ==P"-PG s_1233261703Fp%Rp%ROle GCompObj HiObjInfoJ FMathType 5.0 Equation MathType EFEquation.DSMT49q p& .    &N  & MathType` 3"Times New Roman؟ww wqfr- 2 .y`Times New RomanOlePres000KNEquation Native ]=_1233261704 Fp%Rt%ROle b؟ww wqfr- 2 0.j"Times New Roman؟ww wqfr- 2 s&`FTimes New Roman؟ww wqfr- 2 PPPG*R`Symbol l؟ww wqfr- 2 =-=&-MathTypeUU!DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P Ȱ. ==P"-PG s ==0. & "Systemrqfr !-K! DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P Ȱ. ==P"-PG s ==0. FMathType 5.0 Equation MathType EFEquation.DSMT49q  .  @`&N     "2 !$#)%&'(*0+,-./318G45679>:;<=@?EABCDFMH[IJKLONRPQSYTUVWX\Zao]^_`bhcdefgjimklnvpqrstuxw}yz{|~CompObjciObjInfoeOlePres000f4Equation Native   & MathType-a   "Times New Roman؟ww wf- 2 11`,Times New Roman؟ww wf- 2 5X "Times New Roman؟ww wf- 2 t"Times New Roman؟ww wf- 2 ttteT % "Times New Roman؟ww wf- 2 tj- Times New Roman؟ww wf- 2 H`FTimes New Roman؟ww wf- 2 PPPPTR5` Times New Roman؟ww wf- 2 SC"Symbol h؟ww wf- 2 +js Symbol ؟ww wf- 2 `z 2 -=-=; 2 &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P t++1 "-P t ==P 1 "- "H t "S t ==P t & -"Systemf !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P t++1 "-P t ==P 1 "- "H t "S t ==P t FMathType 5.0 Equation MathType EFEquation.DSMT49q{. .  @@&N_1233261705Ft%Rt%ROle CompObjiObjInfoOlePres000VEquation Native p_1233261706N Ft%Rt%ROle  & MathType` "Times New Roman؟ww wm f/- 2 1yTimes New Roman؟ww wm f/- 2 (1`Times New Roman؟ww wm f/-2 )(1,...,)Xx```"Times New Roman؟ww wm f/- 2 ttFTimes New Roman؟ww wm f/- 2 PPtTC"Symbol ؟ww wm f/- 2 +tSymbol ؟ww wm f/- 2 =+=d&`MathTypeUUTDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P t++1 ==(1++)P t (t==1,...,T) & "System/m f/ !-KT DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P t++1 ==(1++)P t (t==1,...,T) FMathType 5.0 Equation MathType EFEquation.DSMT49qh .  @&NCompObj"iObjInfoOlePres000!#Equation Native  & MathType-{   "Times New Roman؟ww wa f- 2 11`6Times New Roman؟ww wa f- 2 Er "Times New Roman؟ww wa f- 2 t"Times New Roman؟ww wa f-2 tttttKtYcE;c "Times New Roman؟ww wa f- 2 tjG Times New Roman؟ww wa f- 2 H`:Times New Roman؟ww wa f-2 qqqqqQYcEV Times New Roman؟ww wa f- 2 KH"Symbol ؟ww wa f- 2 ++j Symbole Č؟ww wa f- 2 +`2 -=-=-KK 2 +&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  q t++1 "-q t ==q t "- "H t "K t ==q t "-q t++1 Q Kt= & -"Systema f !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  q t++1 "-q t ==q t "- "H t "K t ==q t "-q t++1 Q Kt FMathType 5.0 Equation MathType EFEquation.DSMT49q_1233261707&Ft%Rt%ROle CompObj%(iObjInfoZ .  @&N & MathType- "Times New Roman؟ww we fx- 2 14naTimes New Roman؟ww we fx- 2 14Times New OlePres000')Equation Native _1233261708$0,Ft%Rt%ROle Roman؟ww we fx- 2 4`: Times New Roman؟ww we fx-2 (1,...,)Xx```Times New Roman؟ww we fx- 2 14"Times New Roman؟ww we fx- 2 tt"Times New Roman؟ww we fx- 2 Kt`:Times New Roman؟ww we fx- 2 qqtTTimes New Roman؟ww we fx- 2 QtH"Symbolq /؟ww we fx- 2 +tn?Symbola P؟ww we fx- 2 +t` 2 == 2 +=&MathTypeUU}DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  q t++1 == 1++1++Q Kt q t (t==1,...,T) & -"Systeme fx !-K} DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  q t++1 == 1++1++Q Kt q t (t==1,...,T)CompObj+.iObjInfoOlePres000-/Equation Native       !"#&)*+,-./0123456789:;<=>?@ABCDEFHIJKLMNQTUVWXYZ[\]^_`abcdefghijklmnopqrtuvwxyz} FMathType 5.0 Equation MathType EFEquation.DSMT49q .1  @&N` & MathType "- X j Times New Roman- 2 (,)`Times New Roman- 2 (,)m"Times New Roman- 2 tt<"Times New Roman-2 KtKtttj Times New Roman- 2 QKR`.Times New Roman- 2 QQKRTimes New Roman- 2 Ktj Symbol- 2 ` 2 == 2 &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Q Kt ==Q Kt (K t ,R t )== "Q(K t ,R t )"Kt &  "-"System-pа~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Q Kt ==Q Kt (K t ,R t )== "Q(K t ,R t )"Kt0 FMathType 5.0 Equation MathType EFEquation.DSMT49q_12332617092Ft%Rt%ROle $CompObj14%iObjInfo'OlePres00035(Equation Native G_1233261710*B8Ft%Rt%ROle O* .1  @&N4 & MathTypep "-@ "Times New Roman- 2 11_hFTimes New Roman- 2 0,c 4Times New Roman-2 (0,1...,1)|f```7"Times New Roman- 2 t"Times New Roman- 2 ttRt_; "Times New Roman- 2 tr Times New Roman- 2 HhpTimes New Roman- 2 PqQN6Times New Roman- 2 RTimes New Roman- 2 tTvm"Symbol- 2 ++_rRSymbol- 2 h  2 ==-+:,| 2  2 =-&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H t "R t ==0=="-P t++1 ++q t++1 Q Rt ,(t==0,1...,T"-1) &  "-"System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H t "R t ==0=="-P t++1 ++q t++1 Q Rt ,(t==0,1...,T"-1)CompObj7:PiObjInfoROlePres0009;SEquation Native s FMathType 5.0 Equation MathType EFEquation.DSMT49q .  `&N & MathType- a  z E"Times New Roman؟ww w fv- 2 ,y"Times New Roman؟ww w fv- 2 ,ygM Times New Roman؟ww w fv- 2 ()Times New Roman؟ww w fv- 2 ()"Times New Roman؟ww w fv- 2 tt<"Times New Roman؟ww w fv-2 RtRttt"Times New Roman؟ww w fv- 2 ttg- Times New Roman؟ww w fv- 2 QKR.Times New Roman؟ww w fv- 2 QQKR&Times New Roman؟ww w fv- 2 Rtgs Symbolp ؟ww w fv- 2 t 2 ==l 2 =&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Q Rt ==Q Rt (K t, R t )== "Q(K t, R t )"R t & -"System fv !@-K( DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Q Rt ==Q Rt (K t, R t )== "Q(K t, R t )"R t FMathType 5.0 Equation MathType EFEquation.DSMT49q*~ _1233261711>Ft%Rt%ROle {CompObj=@|iObjInfo~OlePres000?AEquation Native _1233261712<HDFt%Rt%ROle .1  &N@4 & MathTypep "-@ P "Times New Roman- 2 1hFTimes New Roman-2 0'(),64Times New Roman-2 (0,1,...,1)|fx```7"Times New Roman- 2 t "Times New Roman- 2 tt"Times New Roman- 2 tr Times New Roman- 2 HhbTimes New Roman- 2 UCq(F$Times New Roman- 2 CTimes New Roman- 2 tT "Symbol- 2 +rRSymbol- 2 h  2 ==-:bj 2  2 =-6&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H t "C t ==0==U'(C t )"-q t++1 ,(t==0,1,...,T"-1) &  "-"System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H t "C t ==0==U'(C t )"-q t++1 ,(t==0,1,...,T"-1) FMathType 5.0 Equation MathTyCompObjCFiObjInfoOlePres000EGEquation Native dpe EFEquation.DSMT49q  .1  @ &N`  & MathType "-V  jTimes New Roman- 2 d() `Times New Roman- 2 '()6Times New Roman- 2 dO "Times New Roman- 2 t9"Times New Roman- 2 t"Times New Roman- 2 tjTimes New Roman- 2 UC`Times New Roman- 2 UC(Times New Roman- 2 C`Symbol- 2 =&RMathTypeUUFDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  U'(C t )== dU(C t )dC t &  "-"System-pH~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  U'(C t )== dU(C t )dC t FMathType 5.0 Equation MathType EFEquation.DSMT49qj .1  @&No & MathTypep _1233261713JFt%Rt%ROle CompObjILiObjInfoOlePres000KMEquation Native _12332617146fPFt%Rt%ROle  Symbol- 2 { Symbol- 2 }A Symbol- 2 { Symbol- 2 }k Symbol- 2 {  Symbol- 2 }  Symbol- 2 { Symbol- 2 }"Times New Roman- 2 1111*:t"Times New Roman- 2 1111*:"Times New Roman- 2 TTTT*:C"Times New Roman- 2 tttt*:Times New Roman- 2 PqKS64"Symbol- 2 ++--*:&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P t {} 1T++1 q t {} 1T++1 K t {} 1T"-1 S t {} 1T"-1 & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  P t {} 1T++1 q t {} 1T++1 K t {} 1T"-1 S t {} 1T"-1 FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObjORiObjInfoOlePres000QSHEquation Native  O  .  ` &N>  & MathType d,Symbol ؟ww wfm- 2 {ydeSymbol m؟ww wfm- 2 }ydSymbol ؟ww wfm-      !$'()*+,-./012345678:;<=>ADEFGHIJKLMNOPQRSTUVWXYZ[\]^_`acdefghknopqrstuvwxyz{|}~ 2 {ydSymbol n؟ww wfm- 2 }y}_Times New Roman؟ww wfm- 2 11MB_Times New Roman؟ww wfm- 2 00MB` Times New Roman؟ww wfm- 2 60_Times New Roman؟ww wfm- 2 TTMB_Times New Roman؟ww wfm- 2 ttMB/"Times New Roman؟ww wfm- 2 TTUT"Times New Roman؟ww wfm- 2 CRb Times New Roman؟ww wfm- 2 TR2_Symbol h؟ww wfm- 2 --MB: Symbol sl؟ww wfm- 2 =-&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   C t {} 0T"-1 T   R t {} 0T"-1 T  ==6T & "Systemmfm !-E xDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   C t {} 0T"-1 T   R t {} 0T"-1 T  ==6T FMathType 5.0 Equation MathType EFEquation.DSMT49qn{x <.1  @&_1233261715VFt%Rt%ROle "CompObjUX#iObjInfo%OlePres000WY&Equation Native 9_1233261716T`\Ft%Rt%ROle ?N` & MathType` 4"Times New Roman- 2 1Times New Roman-2 '()(0,1,...,1)6 |fx```9"Times New Roman- 2 tt3Times New Roman-2 UCqtT(v"Symbol- 2 +Symbol- 2 ==-6&oMathTypeUUcDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  U'(C t )==q t++1 (t==0,1,...,T"-1) & "System-pd~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  U'(C t )==q t++1 (t==0,1,...,T"-1)CompObj[^@iObjInfoBOlePres000]_CEquation Native b FMathType 5.0 Equation MathType EFEquation.DSMT49qZ .  @&N & MathType- "Times New Roman؟ww wqf- 2 14naTimes New Roman؟ww wqf- 2 14Times New Roman؟ww wqf- 2 4`: Times New Roman؟ww wqf-2 (1,...,)Xx```Times New Roman؟ww wqf- 2 14"Times New Roman؟ww wqf- 2 tt"Times New Roman؟ww wqf- 2 Kt`:Times New Roman؟ww wqf- 2 qqtTTimes New Roman؟ww wqf- 2 QtH"Symbol h؟ww wqf- 2 +tn?Symbol h؟ww wqf- 2 +t` 2 == 2 +=&MathTypeUU}DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  q t++1 == 1++1++Q Kt q t (t==1,...,T) & -"Systemqf !-K} xDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  q t++1 == 1++1++Q Kt q t (t==1,...,T) FMathType 5.0 Equation MathType EFEquation.DSMT49q* l 6.  @ &N  & MathType-_1233261717bFt%Rt%ROle iCompObjadjiObjInfolOlePres000cemEquation Native _1233261718ZrhFt%Rt%ROle     "Times New Roman؟ww wf- 2 1ya"Times New Roman؟ww wf- 2 0y Times New Roman؟ww wf- 2 ()Times New Roman؟ww wf- 2 (1`Times New Roman؟ww wf- 2 )5F _Times New Roman؟ww wf- 2 t)B "Times New Roman؟ww wf- 2 tT"Times New Roman؟ww wf- 2 tT"Times New Roman؟ww wf- 2 tTTimes New Roman؟ww wf- 2 UCTimes New Roman؟ww wf- 2 WCb"Symbol h؟ww wf- 2 =- 2 =-Symbolj h؟ww wf- 2 =- 2 +-aSymbol l؟ww wf- 2 -&MathTypeUUyDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==p U(C t )(1++)  t t==0t==T"-1 " & -"Systemf !-Ky xDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==p U(C t )(1++)  t t==0t==T"-1 "CompObjgjiObjInfoOlePres000ikEquation Native $ FMathType 5.0 Equation MathType EFEquation.DSMT49qqp .1  &Nr & MathType "Times New Roman- 2 1#"Times New Roman- 2 1Times New Roman- 2 ()\ Times New Roman- 2 0"Times New Roman- 2 ttt#S"Times New Roman-2 tttttB:o"Times New Roman- 2 TTVt@Times New Roman- 2 SSRFTimes New Roman-2 KKQKRCx @Times New Roman- 2 SK["Symbol- 2 +# 2 +tSymbol- 2 =- 2 =+-J  2 ==V&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  S t++1 ==S t "-R t K t++1 ==K t ++Q(K t R t )"-C t S T ==K T ==0 & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  S t++1 ==S t "-R t K t++1 ==K t ++Q(K t R t )"-C t S T ==K T ==0 FMathType 5.0 Equation MathType EFEquation.DSMT49q=4 _1233261719nFt%Rt%ROle CompObjmpiObjInfoOlePres000oq\Equation Native ,_1233261720lxtFt%Rt%ROle .1  @&Nx & MathType "- r"Times New Roman- 2 .NTimes New Roman- 2 C@:Times New Roman- 2 rlTimes New Roman- 2 C@zSymbol- 2 rh"@<Symbol- 2 =+p&MathTypeUU DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  r==r++h  C . C &  "-"System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  r==r++h  C . C FMathType 5.0 Equation MathTyCompObjsviObjInfoOlePres000uw Equation Native  pe EFEquation.DSMT49q ^   .1   &N f & MathType v"Times New Roman- 2 0"Times New Roman- 2 .&Ti     !"#$%&'()*+,-./0234567:=>?@ABCDEFGHIJLMNORUVWXYZ[\]^_`abdefgjmnopqrstuvwxyz{}~mes New Roman- 2 ()dTimes New Roman- 2  and`"Times New Roman- 2 tz "Times New Roman- 2 t("Times New Roman- 2 t"Times New Roman- 2 tC)"Times New Roman- 2 tt#&"Times New Roman- 2 to"Times New Roman- 2 t&Times New Roman-2 WUCetO.Times New Roman- 2 CR @Times New Roman- 2 SRzp "Symbol- 2 r+"Symbol- 2 =z 2 - 2 =&Symbol- 2 = 2 =  2 =-,Symbol- 2  &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==1U(C t )e "-rtt==0t==" +" dtC t ==R t  and S .  t =="-R t & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==1U(C t )e "-rtt==0t==" +" dtC t ==R t  and S .  t =="-R t FMathType 5.0 Equation MathType EFEquation.DSMT49qm`` 0.  @&N_1233261721zFt%Rt%ROle CompObjy|iObjInfoOlePres000{}Equation Native 1_1233261722Ft%Rt%ROle 8 & MathType _Times New Roman؟ww wf- 2 0yB*Times New Roman؟ww wf-2 ()max(,,)d(())P&JPP~(_Times New Roman؟ww wf- 2 TyBJ"Times New Roman؟ww wf- 2 ty#>"Times New Roman؟ww wf- 2 Ty^V"Times New Roman؟ww wf- 2 tyLTimes New Roman؟ww wf-2 JuLxuttFxt\z2P2RSymbol `؟ww wf- 2 =+ J^Symbol ؟ww wf- 2 +&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  J(u)==maxqL(x,u,t)dt++F(x(t T )) t 0 t T +" & "Systemf !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  J(u)==maxqL(x,u,t)dt++F(x(t T )) t 0 t T +" FMathType 5.0 Equation MathType EFEquation.DSMT49qEOZ .1  &N@ & MathTypeP c"Times New Roman- 2 CompObj9iObjInfo;OlePres000<Equation Native K,.`@Times New Roman-2 xf(x,u,)L``FTimes New Roman- 2 t`fSymbol- 2 =&MathTypeUU DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x,u,t) & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x,u,t)_1233261723Ft%Rt%ROle PCompObjQiObjInfoS FMathType 5.0 Equation MathType EFEquation.DSMT49qOb .1  &N & MathTypeP c"Times New Roman- 2 OlePres000TEquation Native c0_1233261725Ft%Rt%ROle h.`@Times New Roman-2 xf(x, u,)L```Times New Roman- 2 t`fSymbol- 2 =&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x, u,t) & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x, u,t) FMathType 5.0 Equation MathType EFEquation.DSMT49q`F .1  &N & MathType "-@b `CompObjiiObjInfokOlePres000lEquation Native | Times New Roman- 2 0HTimes New Roman- 2 un Times New Roman- 2 HnRSymbol- 2 ` 2 = 2 &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H"u==0 &  "-"System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H"u ==0 FMathType 5.0 Equation MathType EFEquation.DSMT49q2qr .   &N o & MathType w _1233261726Ft%Rt%ROle CompObjiObjInfob + c $A)? ?3"`?&2Ħ(YxsVf`!^Ħ(YxsV ` ` ,xڕSkA$B>ZQ궠SPz MbzIYLMUI#HIKx.sP^xc=fSKP3= #؆Zaނ(@$yo= g@KC޽[߳eF ͋˴~b|!6Z=ZUދou{oGX"k3#\)#eh&vßG|o*vC%Gy'^G17frcbi9aRCCHE[7M[D;"y2%tMbc#EaBX* 0wi4uq3߾e;dxgR9%NK$Ei.R..ete_-0>7`bݴ+,10V*3q-HSɝDt(lc"4[z\$(}epA5c@ F\rBKӵ[K^3t{@"FGDd b , c $A*? ?3"`?'2tf6[v%^m<`!etf6[v%^    3xڕSKkQI&yTi@BR7NnR2v@H&Vt&݋R.B r#c\ɤh/s8ߜ{;\I@ lB-? !AE)qiji䧑.1PXF-?fy;䈰祡Pm_)uZ@2B 4=*BTkXlusFu |coGڥ%ٛTHT/+0{r?Jn uO*/z`'a1oLWoLLsFL*t/ :>8odV,di: 7XY B 0ufUv1\eF?3&3VjMĨPmխvqgsa:Hl_0>1No((9s%ZXXb`̕圣># ^$%ܶT֟K-仓FZ6-+_ 6R>D|˼Qm]{_zc'op%]m5nB,-c|,aRDd J - C A+? "(2Htg깎`!Htg깎tRxڅR=KA};1j>. h h0A˗6`,S CFKK+E ?J亀Ty;3v ߦ6Ƌ uU$%fUdTCazke2a<]ybkDȫ huO+3+#iN<#hH W= wB9^5N|׾V{*Y փL0#*xl_l6wZ}[FrDkM^XQulm_8l9mkv[&ծ.0siڶi4NYve睤Ud#i̼#?C ! 郒h&- MoRc?s jC1vG0ytdFWHuTDd  J . C A,? ")2@ň,r1(`!@ň,r1(4R(+xڥUkA34e7UĢm@I1$KKɦxLYj i4Y!S@=!*/PIٛfvѬ+M]vwޛ7P~IPB@Ƒv-f䌘L 4BB)5c(])0m3;(g= @#Wl(Gtnq~h[D ll&W{J90#+AƝ"6cٰhD?]s]yox6*Dmw~u6xJrݦaQ}Ւ|;fDev3{tJzFNɟX/۲fN z|Rr26Kd/[;ʿpD/ͧw$8|^&BT'Hv1JgltO ߇-iX-Xkko UMTZ׊70jwU9dnhTt9_- OͦYmRk9mu9 uj,PR$ik:d5PЦs\ⷦA(^3 tTbK >TzĀ\(w@|H?2v޵t`_zK}ķ]' K/^?c)ì:.? `N%N4LWTLLaDd J / C A-? "*2wS ޙ7H o& AS`!KS ޙ7H o& A,@ WxڕSkQf66Mkh XBRTD<MBzI IF=KT&M$7"q!K<'M*Z`QF6$hT +E؁%_l^V,(5j 5KDBg:ƲoiQxN0j;Ai7NJS2/[TpGf(`{lg3'J*M xiJ7QOj=/m<,OjOOx):V㳨fϘo@ךI} 8Gcw97}jws{nq!!nsݦUsmE_, vٷbŚ]z _[5AV,̰oԶd>`o򵂯^;%Jby$!D dM%dxH9jlMOMzR CO3ȸ(Z7aCb)qVf#k nœNI:D9;Dd <b 0 c $A.? ?3"`?+2Y >RJaҏ`!YY >RJ ` ` 'xڕSMkQ=MǤH&"4B )\&1ݤ M$ꨁnPA\"(nܸpǸ|I,hy޻{O`t HNC 1FBCxoINBeBOqC6&LyK +| UJ^s*"ǓN+?elAOI,VgmmkQG.YI vbdg F~/?_5T|8qv8$BG Ԡ k0:c";|&֐ ORHI=@qcb9c>b|4If}]PZ<I1֠fH9axC!*bsܻeUeA䉑|L \oڎfwVwOu6ȚlP8 toy|{f3y'9_7^e`; /a,N\,嘲913nBHddeKwW JwgFŒj8-# dq3d,qP2)e}s%! Orb遉RMş l[{uv"#Dd  b 2 c $A0? ?3"`?-2tWB.u\`!tWB.u `@xڵUoA3mY0G=46JIꡈ &=nBC ހj1P !xƤ/AùLEތEK5ty7o `ެPB VYɄS3A-у&Ńayv FYe5U-zϣq+@tq 4kvn87g D'OT rŻT,dx?*#7></"`pZ':zwwt `i6F,䪽_9L,AiJ `CQ{(L_EB^WM/Gif%C/8::N9'-ŖXl1I4hxؗB8GZX)ywlpC:АΘ-/Y~&uey+Sm9'񄔎T!~6H-LdT3Rr_'ߖ/<7"v;2HFdc/ ^7l\?sCn>#~z}؄ml9} FkE3ۗc9:|<X(>/1zs 9 N15\dR(M>'3QX0KD*X4L~$kSxꗲӧX>a%3Vm2˒WOȥƲVMY&x7B|,;MZ=ĵ WUTjՓY+ńpx ܬX B[8l@ c2.IV3Pp/?l4 ҶŔFf(!"R0RCvv!MDd  b 4 c $A2? ?3"`?/28w"4"S zc`! w"4"S zch@8\ xڵUOA#H4M%5у mhҔSjIK]C0$L/&`84xhB"9U 07 %L;;o7o[`-J3,ZMγF8Yvvu ]P0'Bu'/wQ.Heduhmи-2؅A֦SXvw߱.v_Ӈ?c: Hټ\ 3|f V^Ymd\Wtj/8AмTa:8^j79<:h[Xո {5ԠXIE (US2n8Ju% >9d']$p]U#Q5Wh/8::GiQ͹LE~T%몖1acd6X_\U4${j b]d)X9t}/$Iޘ՗V-#+13+}fVVF/RrRv"Z%e%X%0W{~:6]tDd  b 5 c $A3? ?3"`?02<Tw عBv`!Tw عBvZ@xڭUMLA~3ӥPjR= P,I#IHʪ6i55$B㙃'$7nxdX7o|߼yˠ @ l] XYbT*R-뚹e}.Q:V(15(±e+/- 34'Mn.f0V$MRqsvf(%(OerZ!4͆.s{u=P۬>'yyg;S<=+;-{<5a95Vz=y+p;\uU*ow@#.elZ\_Ӓ :Q# odz >d5]hlI{Iﻐ'wOG/F8i(}(&Q45};B+=N `/taAPTqz 9ѱG/!~35x@eqV- 6,,DϿT;lfdxMHtK):)77Lͱ.dUmjM J|ArTAMh&a|h)v sGqDd  J 6 C A4? "12')H̅Awkw"`!')H̅Awkw"@@xڥMhSAg%3/=XEk@iKPfi,[ I#IQ$K x 9=O^ qm^5d~;39"8yDvf'yi4b̘CWlUPY6 Ow*s@Gߙ2= gΣ!!X %N'Z· ~Н] xjXSڭLR7yy<fL0"9*̱s VxM"(oʵ}V-u}6L}F y%@ 'ʶYlI3l|0If2య.&L[`g v;wz/_stx6;ϯRq+Z./+UmpMC9u@pƪ7kE&!@B:ab_b> 'Q0igILijxBt6' a q`r7^Ǹd6#ܝ=ڜf~6ZbByA᳴ͮvBVU췰u+}Sᐩo6pQ׵J*UL_QDd  J 7 C A5? "22[OFu"7`!/OFu"@ xڵUOA e)Dc4M4@?54@H)%kUUCB" /_?GΞ)k#Kt=gf!uKÚ3| ;szRֆS0)}a+?8cӏew n D;TWk4R=X'^yPYwi45ﱯ̺ff>2MH_Nn>Ϗf˹ٱ}-}'7'1RN+q%p"[k8ƒ5z;K~C[A-^Tfy1>T㵁LV&2xcQ Q@'x-˜,k IM"褢 )C: QC`BZ?D&I= ÈXX2"؞tA٫mwR ;;հg7\V۵h`/ ʺVKAv }Dd  b 8 c $A6? ?3"`?32 SXkLq`!SXkLq`xڭUkAo6Gf=hMo6BII1ZIS/,x0RI̹A׃7lj&Λ7޼AK8ќbCBV i5tnf솣IQ`K*益Uay5dݕ҆P>x~ϰK'*~*C$b(]b`I $:|z }sWsA.p,3{Ylb1Qs71"Ac$L2{ fo …mw*vbs6Qz'xs?g65^u]SrĠ_{Zΐl6&,TYzl@Xf.-5,mRf> !.>sw'"C6ˬ3a㬫u֭F'E+t]Ϭٹt1sk>[״"̨R+dX҅9od&Z̄W ǿ_$=_+s)ʼ*Q}"W ,HO"."#@F"%H#I.eanLʼn6 (-`kLLpzmMZd߲f^ X=#[Kmլz5k v߰T"r(0P"__jֻdz'Dd  J 9 C A7? "42X@=I>2J4`!,@=I>2J~`z xڵUkA&46ւ&m6R+ij.-IS]m$idE A/'"={&*o+mdw~ޛ :[`]nd $uMd4I=B0Wݽ.]pb̂Lw= 0fYfBjeIXb ^-@3t؁%u k1qOuur`X.(?C l1mm:ò6~%V*F>?q}u;3< H&6~q{! Jajgd}ixk\yQޑ֜Áq;]}df'K3'N%gR[]-Fl9w3Z$w2[rJI(3R^QU聎s#H$Q]2ķ\jj !C5VkUM2A8Pf*d$C(!F-fu* z6*q~C" 9} uOG4Y  OLHjAa&ΐeB5Uh*ւ!,gުLowʼ<֙ 4fC$WʪR0O=~VDd J : C A8? "52ҍidш3`s `!ҍidш3`s  @XyZxڭTOkQͶXAmZEACM$zLY4%i$Y"R/`%(b({R1{/mZ̼f޼Ah  F_BW~/5\&YHRBo8K'pbɚ,w]7~ikl}Ph?tz`d\@>BtRXu,T8`u5?3Z$2s$jbN6`*ll7/Tmڄ1㞲促vʳĦ!L7&#邙ߕo CtGGJak5Rc%̳arwQ.RML؁0Y vbY5n8!eEl(qIGPQ/gkg߹U>Do%>f}wʒ:?gyh ST;/Mr|'[]OFMZŐzR9cBL"SىaYHK{L|$a'v'CL2}vYjPx[7`># $U+2T3 ORdA4rẢnCj w l7i_WQEu49-ZMi,-Ä)7EM0l6&Μ3|sfaPnsPKn+-qv^ZQ/)G UA!Ob63Զt;MrȖGcؒ( ,nL̒2)cRphJlo7q"L'pzh{ >)͹g-¡h"ǣ E񜘈=32ae[-6Dz+N0ʉ bj* ğ*W />m\oի`W Wmpjd o=+>( *,O糅ʎQT+?4н+ں7UҨ γFմ,LL'dlMS2r?/]$}V5`g+vgDdҺݑU/I1KRm sFdH~F"u aHE[J|apϫ$d {v 9,У1 Lc޻B@&Hb0jZfmL+2]EDd  b > c $A<? ?3"`?92?Z5lTρ r`!Z5lTρ rZ@xڭUMLA~3ӥPjR= P,I#IDo*MZj`DHz!^LOInܼxDX7ogo2h@3V/bKFnsx(@3i1 5k(X|G*7 %u`Q7pN2~+H$|_f~֢IS5Vg糙{ef G u&\~) ?6?]Bj*N&}+HR}(S7S4JNUM&o#s.9vC(1^Mѣ;z юD"?q34qX`8E^]6pca^ B05D5xZjsŝ7R07ؠD#]RI|oI|'B7-Ij2V7Ǽ̓Y]5VSZf cB 7sCSi2WsDd Tb ? c $A=? ?3"`?:2Z$g5Jg>"T6`!.$g5Jg>"Tl @xڭUMLA~3jb$Zh(&xkB)U."m8EWi)kHOVpg/ƛO&8EL$^ 7;[ Ԩy3 %BV*!,Y;)t;n@'Tb/JV?g >2V)03œ m(E&7P~Q57GnMvv\`E' x鬖k : .'VJ=7(,{ O8%" ݿ ;:g4]?u#} 빐N|9yOq(y=1>FOj(!o"FRq`\ j4DxEJI| guTc.$k^p37}Ck7ks=跈RC0l! n-jˣs +^h5 LHp׵,t|o3/i vADd ` 8J @ C A>? ";2Í9J`!Í9Jp@v xڥUkA~M1t7R E4MfzPQ $$+ړ-Ƀ=z=kD ofI˼yoy3 B/HJEyn%'$UOe !nh:\`GU'_X;(Q%Ik;&+{ōIL*NZ&\z8a2xE^P?A->u{.**=5HpqWNl砆z6Ek:*' ض 55{n0^я[" ;f: ЪưU1jݞ:6A_5ʕ߬lyF!kn糀3/x}N}6ўgӉu8tpJԼ'`~}(o'oaJ}K̕mH6.fDO3oQF/  ;!ũ=}WgcȚcR_W#ĜNi1ŖGeU93e< \Yf#2ZJEt?H\xAB Pbo޼kPxt,0 CiuiFAOV9XLgnpB :`~<#P.CۙC|8NДkn 'UdX}- Dax'"ȝ;Qxۤ񚷳U8gB6(܅KT?z9lhX0` =L?NUvWκH*goШPkPu<XL}sS7Z mu]F[|&e ` R. knڭ/K J@!AT@.+A 7t*+/A!\=9yUe;EwqFMiw%L+:Ʃ0l2uGI()+٩Dd xJ B C A@? "=2wy//ܐ62.?e ci `!?e ci`@H xڵTMhA~ovӟ44X6MҴM--M"ѮHm64"^"AV&=x(O"1^*LRĊ73oo@Q?D(Ţ9$v6ꨳ l ࠙.όy  6MxqSK;p7*K;*RȬI#ɖ㨭1:rb+c:W8 F)NgqREPrbB١.'3swTh.4YʴcG-675YjwA])OC XX@˙/$3z0rIOR_"w &RsQ})"s10ں١n_ట Ç}PaB wY3ͨi"Cƀ9TW%8O+-+pN3p`yidh{=nF{t<7Gk]Oi$ =\\gmp ՕKSb$$If!vh55#v#v:V l4055/ 4f4Dd J D C AB? "?2OQċF_%2֡+8`!#QċF_%2֡Z xڍRjQΙILZDcAiK2i”`R\N:@H2deJHН/ ]|AdvwӅ^9~`|&qfd3"0Ԛ-k}W10䢠{&n"TK4Oeܘ25˜c>}’pLZ0f r;ޛ^=3Ti妜 8mK1e#-֔Qkuמ{hv:3(M㒵EY6h:뿹uw~.:{\iԚgrrWw_zdәg8j6Q mrw<Rw6ʦ],cܾ%xɴTdde%&`&ZSv[Tj>̐A,Kj&)Ӏ.5Ѝ0F2]Ʃߎ Rmg>QSKV#'9*B}T.Sp4$@{ Iw^Yi׽ p$$If!vh55F 5j #v#vF #vj :V l4055F 5j / / / /  / 4f4$$If!vh55F 5j #v#vF #vj :V l4055F 5j / 4f4$$If!vh55555R#v#v#v#v#vR:V l055555R/ 4$$If!vh55#v#v:V l4055/ 4f4Dd J E C AC? "@2SbvvTXk//]`!'bvvTXk/b6 xڍRn@f4i)NKP1@mR1R$RE$G"% J(Eczpʝ+C 3vI|;3S1'Igqh øH. ;Mva\T$0W5fYP|5> ΫN 'l^[Oz]0إ F # AD"A2tra%1YO^N`!Vtra%1YO h p$xڍS͊AO62"=,&ǀI0,dGw$$brwS ^|@d݃DX՝!gd꯿ Wa0T8,PЗ(_dz!E‚ !M Yvz+IQ ({}#kėR mufFhF2Jç-v ^VmaFe(?>r88 GB {C=y\6T9>Vģ9V_gXK-<ÊZ4o}VO)$7]S"rbJ0wLGL:cg1jD,T˵t7-nNUܻ^%Xo2o?4;!e{ҹ⨰T˼^w*Ξ!f:hN2Rq4s瞧68pCӾ2ݘɰ7jjWݞלtC_["$$If!vh55^5R#v#v^#vR:V l4055^5R/ / / /  / 4f4Dd  lJ G C AE? "B2Eq!soB9!V`!q!soB9pxڕRKK@M}Z=⣠'x,b.b+cԠ>'xѣBBx3L6oGWb*xMQ}&g .)7PڧoUXٟ9j.劂5]os~;zmꞖokkOx:Y ֳ=[٭XqNY;vBC6]¤Uٕ ALS.zydpƝƒIHSc4Τi8 Y${`'͸pħAFe0Eb=Q9Bh;*Q"BDwG\_SW/ :FP"t\s_{.> & $$IfOlePres000Equation Native _1233261727Ft%Rt%ROle Times New Roman؟ww wf-2 ()0 if `l~4Times New Roman؟ww wf-2 or ()0 otherwise~t`l~l Times New Roman؟ww wf-2 ()'()N6"Times New Roman؟ww wf- 2 TTc"Times New Roman؟ww wf- 2 TTwTimes New Roman؟ww wf- 2 ttTimes New Roman؟ww wf- 2 FtTimes New Roman؟ww wf- 2 tF8wSymbolO A؟ww wf- 2 lF 2 lFwSymbolS ؟ww wf- 2 ==A&^ 2 =, 2 =$&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  l(t T )==0 if t T =="or F(")==0 otherwisel(t T )==F'(") & "Systemf !-E DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  l(t T )==0 if t T =="or F(")==0 otherwisel(t T )==F'(") FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObjiObjInfoOlePres000\Equation Native {4 .1  @@&N & MathType` Times New Roman- 2 ()0r"Times New Roman- 2 TFTimes New Roman- 2 HtSymbol- 2 =&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  H(t T )==0 & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  H(t T )==0 FMathType 5.0 Equation MathType EFEquation.DSMT49q{D "_1233261728Ft%Rt%ROle CompObjiObjInfoOlePres000lEquation Native _1233261729Ft%Rt%ROle   ."System0 -@Times New Roman- ,--.  @Times New Roman-. dhTimes New Roman- 2 dh()0n~Times New Roman- 2 ~TsdTimes New Roman- 2 dHtwdSymbol- 2 d="System------'"Systemm|'m0-''' 2 d --p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  H(t T )==0 FMathType 5.0 Equation MathTyCompObjiObjInfoOlePres000$Equation Native pe EFEquation.DSMT49q` ~.1  &N@ & MathType _Times New Roman- 2 0B*Times New Roman-"2 (u)max(x,u,)d(x())&`~(_Times New Roman- 2 TBV"Times New Roman- 2 t"Times New Roman- 2 tT #9"Times New Roman- 2 T^b"Times New Roman- 2 tLTimes New Roman-2 JLtetFte ]"Symbol- 2 rr  "Symbol- 2 -- ^Symbol- 2 =+] JjSymbol- 2 &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  J(u)==maxqL(x,u,t)e "-rt dt++F(x(t T )) t 0 t T +" e "-rT & "System-p~l~DSMT5WinAllBasicCodePages                       ! $ ' ( ) * + , - . / 0 1 2 3 4 5 6 7 8 9 : < = > ? @ A D G H I J K L M N O P Q R S T U V W Y Z [ \ _ b c d e f g h i j k l m n o p q r s u v w x y |  Times New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  J(u)==maxqL(x,u,t)e "-rt dt++F(x(t T )) t 0 t T +" e "-rT_1233261730Ft%Rt%ROle  CompObj iObjInfo  FMathType 5.0 Equation MathType EFEquation.DSMT49q,OP (  ."System0 j - @Times New Roman-,-OlePres000 xEquation Native  ,_1233261731Ft%Rt%ROle " -.  @Times New Roman-. \&Times New Roman- 2 \&.tTimes New Roman-2 xf(x,u,)0&25ETimes New Roman- 2 t]Symbol- 2 ]="System------'"Systemm| j j/ jm0-''' 2  --p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x,u,t) FMathType 5.0 Equation MathType EFEquation.DSMT49q .1  ``&CompObj# iObjInfo% OlePres000& @Equation Native ; N  & MathType` Times New Roman-!2 e(x,u,)f(x,u,),()&`~~`~"Times New Roman- 2 tt,u"Times New Roman- 2 cFTimes New Roman-2 HHLtteN\vh"Symbol- 2 rr,2Symbol- 2 mmlHv\Symbol- 2 ==+=\Z&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  H c ==He rt ==L(x,u,t)++mf(x,u,t),(m==le rt ) & "System-pФ~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  H c ==He rt ==L(x,u,t)++mf(x,u,t),(m==le rt ) FMathType 5.0 Equation MathType EFEquation.DSMT49qOT *  ."System_1233261732Ft%Rt%ROle B CompObjC iObjInfoE OlePres000F |Equation Native X 0_1233261733Ft%Rt%ROle ] 0' - @Times New Roman-,--.  @Times New Roman-. \#Times New Roman- 2 \#.tTimes New Roman-2  xf(x, u,)0&25ETimes New Roman- 2 t]Symbol- 2 ]="System------'"Systemm|''^'m0-''' 2  --p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x, u,t) FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObj^ iObjInfo` OlePres000a Equation Native t H F G.1  &N & MathType "- "Times New Roman- 2 .6Times New Roman- 2 x"Times New Roman- 2 cjTimes New Roman- 2 H`@Symbol- 2 mrmjSymbol- 2 ` 2 =-H| 2 &8MathTypeUU,DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   m . ==rm"- "H c "x &  "-"System-p,~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   m . ==rm"- "H c "x FMathType 5.0 Equation MathType EFEquation.DSMT49qGF d   @."System@_1233261734Ft%Rt%ROle z CompObj{ iObjInfo} OlePres000~ Equation Native  _1233261735Ft%Rt%ROle   0. @- @Times New Roman-?,--.  @Times New Roman-.- Times New Roman- 2 0UTimes New Roman- 2 Uu_FTimes New Roman- 2 _FH_Symbol- 2 _ 2 =% 2 %-"System------'"Systemm|//.m0-''' 2 E? --p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H"u==0CompObj iObjInfo OlePres000 Equation Native   FMathType 5.0 Equation MathType EFEquation.DSMT49qq .  @&No & MathType wVTimes New Roman؟ww w!fh-2 ()0 if `l~4     ( !&"#$%)'.@*+,-/201437568;9:=<B>?AWCFDEGHILJKMPNOQRSVTUX\nYZ[^]a_`bfcdeghiljkmropqtsxuvwy|z{}~Times New Roman؟ww w!fh-$2 or if ()0 otherwiseY~`l~t`l~lVTimes New Roman؟ww w!fh-2 ()'()N68"Times New Roman؟ww w!fh- 2 TTc8"Times New Roman؟ww w!fh- 2 TTwTimes New Roman؟ww w!fh- 2 ttTimes New Roman؟ww w!fh- 2 FtTimes New Roman؟ww w!fh- 2 tF8wZSymbol5 h؟ww w!fh- 2 mFZ 2 mFwSymbol }h؟ww w!fh- 2 ==& 2 =, 2 =$&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  m(t T )==0 if t T =="or if F(")==0 otherwisem(t T )==F'(") & "Systemh!fh !-E xDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  m(t T )==0 if t T =="or if F(")==0 otherwisem(t T )==F'(") FMathType 5.0 Equation MathType EFEquation.DSMT49qg Z_1233261736Ft%Rt%ROle  CompObj iObjInfo OlePres000 Equation Native  P_1233261737Ft%Rt%ROle  .   &N & MathType jPTimes New Roman؟ww wf- 2 ()8Times New Roman؟ww wf- 2 /0"Times New Roman؟ww wf- 2 cTujTimes New Roman؟ww wf- 2 Ht@Times New Roman؟ww wf- 2 FTjSymbol h؟ww wf- 2 +T4 2 =ͬ&@MathTypeUU4DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  H c (t T )++"F/"T==0 & "Systemf !-E4 xDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  H c (t T )++"F/"T==0 FMathType 5.0 Equation MathType EFEquation.DSMT49qq ].  &N# & MathType CompObj iObjInfo OlePres000 Equation Native  UTimes New Roman؟ww wf- 2 ()Times New Roman؟ww wf- 2 /0"Times New Roman؟ww wf- 2 cTuTimes New Roman؟ww wf- 2 Ht@Times New Roman؟ww wf- 2 fT^SymbolK h؟ww wf- 2 +T4 2 =o^&EMathTypeUU9DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  H c (t T )++"f/"T==0 & "Systemܓf !-E9 xDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  H c (t T )++"f/"T==0_1233261738Ft%R%ROle  CompObj iObjInfo                    ! " # $ % & ' ( * + , - 0 3 4 5 6 7 8 9 : ; < = > ? @ A B C D E F H I J K L O R S T U V W X Y Z [ \ ] ^ _ ` a b c d e f g h i k l m n o p s v w x y z { | } ~   FMathType 5.0 Equation MathType EFEquation.DSMT49qOZ .1  &N & MathTypeP o"Times New Roman- 2 .`@Times New Roman-OlePres000 Equation Native  ,_1233261739>F%R%ROle  2 xf(x,u,)L~~``Times New Roman- 2 t`fSymbol- 2 =&MathTypeUU DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x,u,t) & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x,u,t)CompObj iObjInfo OlePres000 jEquation Native )  FMathType 5.0 Equation MathType EFEquation.DSMT49qB .1  &Nf & MathType0 o"Times New Roman- 2 .`@Times New Roman-2 xdx/dLl`8Times New Roman- 2 t`fSymbol- 2 =& MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==dx/dt & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==dx/dt FMathType 5.0 Equation MathTy_1233261740F%R%ROle . CompObj/ iObjInfo1 pe EFEquation.DSMT49q ` ~.1   &N  & MathType _Times New Roman- 2 0B*Times New Roman-2 (u)(x,u,)d`(OlePres0002 $Equation Native G p_1233261741F%R%ROle M _Times New Roman- 2 TB"Times New Roman- 2 t^"Times New Roman- 2 tLTimes New Roman- 2 JLtt>^Symbol- 2 =JSymbol- 2 &_MathTypeUUSDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  J(u)==qL(x,u,t)dt t 0 t T +" & "System-pT~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  J(u)==qL(x,u,t)dt t 0 t T +" FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObjN iObjInfoP OlePres000Q 8Equation Native j L` .1   &N  & MathType _Times New Roman- 2 0B*Times New Roman-2 (u)(x,u,)ed`+(_Times New Roman- 2 TB"Times New Roman- 2 t "Times New Roman- 2 t^"Times New Roman- 2 tLTimes New Roman- 2 JLtt> "Symbol- 2 r "Symbol- 2 -^Symbol- 2 =JSymbol- 2 &MathTypeUUuDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  J(u)==qL(x,u,t)e "-rtt 0 t T +" dt & "System-px~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  J(u)==qL(x,u,t)e "-rtt 0 t T +" dt FMathType 5.0 Equation MathType EFEquation.DSMT49q1` ~.  &N@ & MathType _1233261742F%R%ROle q CompObjr iObjInfot OlePres000u $Equation Native  _1233261743F%R%ROle   _Times New Roman؟ww w f?- 2 0yB*Times New Roman؟ww w f?-2 (u)max(x,u,)ed&`f(&_Times New Roman؟ww w f?- 2 tBf"Times New Roman؟ww w f?- 2 t"Times New Roman؟ww w f?- 2 t^X"Times New Roman؟ww w f?- 2 tLTimes New Roman؟ww w f?- 2 JLtt)"Symbol '؟ww w f?- 2 r "Symbol n؟ww w f?- 2 -^Symbol (؟ww w f?- 2 =J`Symbol o؟ww w f?- 2 &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  J(u)==maxqL(x,u,t) t 0 t t +" e "-rt dt & "System? f? !-K DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  J(u)==maxqL(x,u,t) t 0 t t +" e "-rt dt FMathType 5.0 Equation MathType EFEquation.DSMT49qOZ .1  &N & MathTypeP oCompObj iObjInfo OlePres000 Equation Native  ,"Times New Roman- 2 .`@Times New Roman-2 xf(x,u,)L~~``Times New Roman- 2 t`fSymbol- 2 =&MathTypeUU DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x,u,t) & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x,u,t) FMathType 5.0 Equation MathType EFEquation.DSMT49qF .1  `&N  & MathType "_1233261744F%R%ROle  CompObj iObjInfo OlePres000 8Equation Native  _1233261745F%R%ROle  -@ `fTimes New Roman- 2 0Times New Roman- 2 u;"Times New Roman- 2 cj Times New Roman- 2 HjRSymbol- 2 `@ 2 = 2 &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H c "u==0 &  "-"System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H c "u==0 FMathType 5.0 Equation MathType EFEquation.DSMT49qOZ CompObj iObjInfo OlePres000 Equation Native  ,.1  &N & MathTypeP c"Times New Roman- 2 .`@Times New Roman-2 xf(x,u,)L~~``Times New Roman- 2 t`fSymbol- 2 =&MathTypeUU DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x,u,t) & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x,u,t) FMathType 5.0 Equation MathType EFEquation.DSMT49q F G_1233261746 F%R%ROle  CompObj   iObjInfo OlePres000   Equation Native  H_1233261747&F%R%ROle  .1  &N & MathType "- {"Times New Roman- 2 .6Times New Roman- 2 x"Times New Roman- 2 cjTimes New Roman- 2 H`@Symbol- 2 mrmjSymbol- 2 ` 2 =-H| 2 &8MathTypeUU,DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   m . ==rm"- "H c "x &  "-"System-p,~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   m . ==rm"- "H c "x                      " % & ' ( ) * + , - . / 0 1 2 3 4 5 6 8 9 : ; < = @ C D E F G H I J K L M N O P Q S T U V W Z ] ^ _ ` a b c d e f g h i j k l m o p q r s v y z { | } ~   FMathType 5.0 Equation MathType EFEquation.DSMT49q  .1   &N ` & MathType >W"Times New Roman- 2 CompObj iObjInfo OlePres000 >Equation Native  0Times New Roman-2 (,,)edf_Times New Roman- 2 tBA"Times New Roman- 2 tt'Times New Roman- 2 Lxutz_Symbol- 2 rB="Symbol- 2   2 -*@Symbol- 2 &rMathTypeUUfDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  qL(x t ,u t ,) 0" +" e "- rt dt & "System-ph~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  qL(x t ,u t ,) 0" +" e "- rt dt_1233261748F%R%ROle  CompObj! iObjInfo#  FMathType 5.0 Equation MathType EFEquation.DSMT49qp L.1    &N & MathType` o"Times New Roman- 2 OlePres000$ Equation Native 7 _1233261749 F%R%ROle > . "Times New Roman- 2 00T`Times New Roman-2 (,) and ()`"Times New Roman- 2 tt`LTimes New Roman-2 xfxuxtx|b2`TSymbol- 2 ==+ &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x t ,u t ) and x(t 0 )==x 0 & "System-pА~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x t ,u t ) and x(t 0 )==x 0 a FMathType 5.0 Equation MathTyCompObj? iObjInfoA OlePres000B Equation Native R Dpe EFEquation.DSMT49qp .1   &N & MathType` c"Times New Roman- 2 .`Times New Roman- 2 (,)"Times New Roman- 2 tt`LTimes New Roman- 2 xfxu|b`TSymbol- 2 =&3MathTypeUU'DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x t ,u t ) & "System-p(~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x . ==f(x t ,u t )_1233261750"F%R%ROle X CompObj!$Y iObjInfo[  FMathType 5.0 Equation MathType EFEquation.DSMT49q p8 .1    &N@  & MathType` {"Times New Roman- 2 OlePres000#%\ `Equation Native n d_12332617512(F%R%ROle t .`Times New Roman- 2 /"Times New Roman- 2 tct`Times New Roman- 2 HxZ`@Symbol- 2 mrm`Symbol- 2 =-Z&TMathTypeUUHDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   m . ==rm t "-"H c /"x t & "System-pH~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   m . ==rm t "-"H c /"x t FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObj'*u iObjInfow OlePres000)+x Equation Native  T <  o.1  `&N ` & MathType >W"Times New Roman- 2 0ATimes New Roman- 2 ()ed+"Times New Roman- 2 t"Times New Roman- 2 tTimes New Roman- 2 UCt"Symbol- 2 r="Symbol- 2 l 2 -*@Symbol- 2 &CMathTypeUU7DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  qU(C t0" +" )e "-rt dt & "System-p8~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  qU(C t0" +" )e "-rt dt FMathType 5.0 Equation MathType EFEquation.DSMT49qy p _1233261752.F%R%ROle  CompObj-0 iObjInfo OlePres000/1 Equation Native  D_1233261753,84F%R%ROle  .1    &N@  & MathType` '"Times New Roman- 2 .`Times New Roman- 2 ()"Times New Roman- 2 tt:`FTimes New Roman- 2 KQKCL`Symbol- 2 =-&3MathTypeUU'DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K . ==Q(K t )"-C t & "System-p(~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K . ==Q(K t )"-C t FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObj36 iObjInfo OlePres00057 Equation Native  tp {.1   &N` & MathType` '"Times New Roman- 2 .`Times New Roman-2 ()(())();:*"Times New Roman-2 tttttt;:5`FTimes New Roman-2 HCtUCQKCUCK+LR`V Symbol- 2 mmA`4Symbol-2 =+-=-R&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  HCt==U(C t )++m t (Q(K t )"-C t )==U(C t )"-m t  K . & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  HCt==U(C t )++m t (Q(K t )"-C t )==U(C t )"-m t  K . FMathType 5.0 Equation MathTy_1233261754:F%R%ROle  CompObj9< iObjInfo pe EFEquation.DSMT49q p q.1   &N` & MathType` c"Times New Roman- 2 .`Times New Roman- 2 // "OlePres000;= Equation Native  _1233261755@F%R%ROle  Times New Roman-2 tcttttt8`Times New Roman- 2 HKQKZ`@Symbol-2 mrmrmm `Symbol-2 =-=-ZW&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   m . ==rm t "-"H c /"K t ==rm t "-m t "Q t /"K t & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   m . ==rm t "-"H c /"K t ==rm t "-m t "Q t /"K tCompObj?B iObjInfo OlePres000AC xEquation Native  t FMathType 5.0 Equation MathType EFEquation.DSMT49qpP (.1    &N`  & MathType` o"Times New Roman- 2 .`*Times New Roman-                       ! " # $ % & ' ) * + , - 0 3 4 5 6 7 8 9 : ; < = > ? A B C D G J K L M N O P Q R S T U V W X Z [ \ ] ` c d e f g h i j k l m n o p q s t u v w z } ~   2 /(/)"Times New Roman- 2 ttt`ITimes New Roman- 2 QK`@Symbol- 2 mmr`Symbol- 2 =-p&dMathTypeUUXDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   m . /m t ==r"-("Q t /"K t ) & "System-pX~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   m . /m t ==r"-("Q t /"K t ) FMathType 5.0 Equation MathType EFEquation.DSMT49q   _1233261756FF%R%ROle  CompObjEH iObjInfo OlePres000GI @Equation Native ( x_1233261757DPLF%R%ROle . .1   &N ` & MathType >"Times New Roman- 2 0Times New Roman- 2 ()edf+ "Times New Roman- 2 t "Times New Roman- 2 t_"Times New Roman- 2 t>&"Times New Roman- 2 tTimes New Roman- 2 WUCt "Symbol- 2 rb"Symbol- 2 = 2 ->| 2 =Symbol- 2 =*~Symbol- 2 &fMathTypeUUZDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==qU(C t ) t==0t==" +" e "-rt dt & "System-p\~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==qU(C t ) t==0t==" +" e "-rt dt FMathType 5.0 Equation MathTyCompObjKN/ iObjInfo1 OlePres000MO2 XEquation Native @ <pe EFEquation.DSMT49qO0 .1   &N & MathTypeP c"Times New Roman- 2 .`@Times New Roman-2 x = f(x, u, )``~~````Times New Roman- 2 t&,MathTypeUU DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x .  = f(x, u, t) & "System-p ~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   x .  = f(x, u, t) FMathType 5.0 Equation MathType EFEquation.DSMT49q_1233261758RF%R%ROle E CompObjQTF iObjInfoH p .1   `&N  & MathType` '"Times New Roman- 2 .w&"Times New Roman- 2 t"Times New Roman- 2 t`@OlePres000SUI Equation Native Y 0_1233261759JbXF%R%ROle ^  Times New Roman- 2 SR`Symbol- 2 =-,&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S .  t =="-R t & "System-p~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S .  t =="-R t FMathType 5.0 Equation MathTyCompObjWZ_ iObjInfoa OlePres000Y[b Equation Native r Hpe EFEquation.DSMT49qfp .1   &N` & MathType` '"Times New Roman- 2 .qt"Times New Roman- 2 t`"Times New Roman- 2 tt`FTimes New Roman- 2 KQC`?Symbol- 2 =-&6MathTypeUU*DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K .  t ==Q t "-C t & "System-p,~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K .  t ==Q t "-C t_1233261760^F%R%ROle x CompObj]`y iObjInfo{  FMathType 5.0 Equation MathType EFEquation.DSMT49q 2 .1   &N@ 5 & MathType "-@@@O "OlePres000_a| ZEquation Native  _1233261761\hdF%R%ROle   Times New Roman- 2 ,tTimes New Roman- 2 0_Times New Roman- 2 tB7"Times New Roman- 2 C"Times New Roman- 2 Ct/i"Times New Roman- 2 tq Times New Roman- 2 HTimes New Roman- 2 UnTimes New Roman- 2 CqRSymbol- 2  2 =-= 2 &vMathTypeUUjDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H C t  "C t ==U C,t "- t ==0 &  "-"System-l~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H C t  "C t ==U C,t "- t ==0 FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObjcf iObjInfo OlePres000eg Equation Native  X ,.1  `&N 5 & MathType "-@@@O A "Times New Roman- 2 t,9Times New Roman- 2 0)_Times New Roman- 2 tB7"Times New Roman- 2 Cr"Times New Roman- 2 tRt-/W"Times New Roman- 2 tq Times New Roman- 2 HTimes New Roman- 2 PQTimes New Roman- 2 RqRSymbol- 2  2 =-+=,D 2 &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H C t  "R t =="-P t ++ t Q R,t ==0 &  "-"System-А~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H C t  "R t =="-P t ++ t Q R,t ==0 FMathType 5.0 Equation MathType EFEquation.DSMT49q }.1  &_1233261762jFP%RP%ROle  CompObjil iObjInfo OlePres000km "Equation Native  _1233261763VpFP%RP%ROle  N5 & MathType "-@C@R s"Times New Roman- 2 .. Times New Roman- 2 _Times New Roman- 2 tB:"Times New Roman- 2 C>"Times New Roman- 2 tt  "Times New Roman- 2 tt/Q"Times New Roman- 2 tqTimes New Roman- 2 HFTimes New Roman- 2 PPPP( pTimes New Roman- 2 SqUSymbol- 2  2 =-+=,k@ 2 &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P .  t =="- "H C t  "S t ++P t ! P .  t ==P t &  "-"System-~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P .  t =="- "H C t  "S t ++P t ! P .  t ==P t FMathType 5.0 Equation MathType EFEquation.DSMT49q !vh55#v#v:V l4055/ 4f4Dd  8b H c $AF? ?3"`?C2PUBېY,`!$UBېYr Pv xڵUMhA~ov7ڸVP 6hP[DKmKk1)ތnm`ӕdeE[+Ez'OTy{[O"TADSnvcrag{{3-K&&Qg5 YZuxvyv2>ɭ$uB*71JԟFjV2m `WZژ2涜p:p&_Jʘf)WBn=i}(4!)|&ȝPPS <-\]bUWEV%B&1T( <6gƃ㑜,tRƄo5xÿx93ei]@(uV-> :(K2oT/o9FLRf}7& bal/&Qҳp9T>DDOՙİ%ӷ᧥uA$+ 5&lz$mq6~Y;wZML|D Ab9֫UsqSJv("D!N1"l4W5Ӥ`g<ܯХ:!}.fI TN}\ j_v8d<"F%"1+Agp < BtdWÔ%.Wj|(N)pF@ɇi>h',?}q#o;d "&G鍧gKV ' K ľDa"8߀4yԺ$$If!vh55F 5j #v#vF #vj :V l4055F 5j / 4f4Dd hJ I C AG? "D27z&d]E-`! z&d]E-4@@Э |xڅRKA$61t7Z/PY\"n#ɖSCx_MpfvDA}߼y8Jf߶v{뷎~QH^0[ rS& \zm?0G[QenƵ La5\ G; ,+3n-iOI엚{íP}'IXLjd&k LKIU*{"۸w;f [ҫ rJs2t> Dd hJ J C AG? "E27z&d]E-v`! z&d]E-4@@Э |xڅRKA$61t7Z/PY\"n#ɖSCx_MpfvDA}߼y8Jf߶v{뷎~QH^0[ rS& \zm?0G[QenƵ La5\ G; ,+3n-iOI엚{íP}'IXLjd&k LKIU*{"۸w;f [ҫ rJs2t> $$If!vh555|55R#v#v#v|#v#vR:V l0555|55R/ 4Dd  J K C AH? "F2T ǹ┲)0- `!( ǹ┲)Ȉ'xڭUOA~vk[(tˏ(HZ&k d+d ly\smMH'[N-_ك5y߾E-279)r8ұ#%}F^7ds$3)- T$71hxf 0R#E̞踙-CT'F#&^0G13 Sjeg8 ʛ0ugYra?Y)ІasN>tB[)%hpX=^%8VA-uKpJ ¸26at@M|@9Lj_vNQ/Mv%m/ $$If!vh5n5#vn#v:V l40!5n5/ 4f4Dd J L C AB? "G2OQċF_%2֡+`!#QċF_%2֡Z xڍRjQΙILZDcAiK2i”`R\N:@H2deJHН/ ]|AdvwӅ^9~`|&qfd3"0Ԛ-k}W10䢠{&n"TK4Oeܘ25˜c>}’pLZ0f r;ޛ^=3Ti妜 8mK1e#-֔Qkuמ{hv:3(M㒵EY6h:뿹uw~.:{\iԚgrrWw_zdәg8j6Q mrw<Rw6ʦ],cܾ%xɴTdde%&`&ZSv[Tj>̐A,Kj&)Ӏ.5Ѝ0F2]Ʃߎ Rmg>QSKV#'9*B}T.Sp4$@{ Iw^Yi׽ p$$If!vh5n5~ 5 #vn#v~ #v :V l40!5n5~ 5 / / / /  / 4f4$$If!vh5n5~ 5 #vn#v~ #v :V l40!5n5~ 5 / 4f4$$If!vh5n5f555#vn#vf#v#v:V l0!5n5f55/ 4$$If!vh5n5#vn#v:V l40!5n5/ 4f4Dd ||J M C AI? "H2})bݯj-+?ran'6L`AjC};JwaWzWVfF7lFD'BDc ]yeA,^}J0`퐻0^Vp$XNڶcٔ)ij C6C,_P&3>msphZm:ȯCrT3;?dp8I붏v`V'][{Cd<4Ifox"J^.EtZ<^o7hQ+5 n,*E-[HP\kWQ"`čFk%rъ;(%P,c߈I=g2Et\L B(a'oAbžY-*f5ѱ #*^9xO!>5kE-A"CiA1)8gF? H{[ d^faw@f]We$$If!vh5n5#vn#v:V l40!5n5/ 4f4Dd J N C AC? "I2SbvvTXk//`!'bvvTXk/b6 xڍRn@f4i)NKP1@mR1R$RE$G"% J(Eczpʝ+C 3vI|;3S1'Igqh øH. ;Mva\T$0W5fYP|5> ΫN 'l^[Oz]0إLvXD zQLM$47)xchJ$b?^‚Ы  R0IV(η;3 5!!3xYG{%9ύ&d*Ů}+[7};_M*HpV9_?Be2$$If!vh5n55#vn#v#v:V l40!5n55/ / / /  / 4f4Dd  lJ P C AE? "K2Eq!soB9!I"`!q!soB9pxڕRKK@M}Z=⣠'x,b.b+cԠ>'xѣBBx3L6oGWb*xMQ}&g .)7PڧoUXٟ9j.劂5]os~;zmꞖokkOx:Y ֳ=[٭XqNY;vBC6]¤Uٕ ALS.zydpƝƒIHSc4Τi8 Y${`'͸pħAFe0Eb=Q9Bh;*Q"BDwG\_SW/ :FP"t\s_{.> & $$If!vh5n5#vn#v:V l40!5n5/ 4f4 Dd 8b Q c $AK? ?3"`?L2W[GكidNOO3%`!+[GكidNOO@v xڵUMLA~ovJmhh$4#KJ!R4rI˚v͆ b㥞SIg,Ϛ<ތ]W^^dƚֺ/(nOHtLSNL1&|d Vв ]UxK-PkzcWesay_>%(! Kts계{ c}a)i%R1 \2F\ +,eISG͹dɷ PL5EhL7$:]ٽ'7WVEpT>]&7bQ.%k\@5Zb7J6e.l^ A۩h"g} ӧ%`3f{KJNsٶ4q3 sT!m?'J?q!XL(aIR]ylȩMf!ԝ"u]ù&a 1[.߂A~[~VЫؕ(p%x>R,@IdDd 8b S c $AM? ?3"`?N2b?ܻgs-`!b?ܻgsv PxڭTkAn1UA,F,m<&œ nl iz24PK/yACoQEדfHTD겳yo7,aPn @`x"v6Hcµ󉦺e4c ]iyxeSMpN]٢3}ˈ'$:.Z%}8xDQZWO65=&m&G:wf*.;Ҝ"]1 ,kCn^b8/dY1اC)giсdɯWo*uau诱Y>2Sxt:ڢ8ɧvz3Mno.,\%T>_FoW\^^J]67%kkecU*j0MptpdmV8qYٲjczWTj>;!)XLR]jZya eS_NRmޕPSKc'9 L@7WSp<$@{ \Ͻ${ FoDd J U C AO? "P27W2ࡦ6ɶ;+5`! W2ࡦ6ɶ;BL @CxڍRkQf4"BXC /nL &㺚6ݒl19*(/=C ]JMطyxCXr `ALC$q$ZS{ZUEF 1f=5D],DȎ@n,f^4}g|O4˂c9X {Ҟ2軇YbJ,ܔs甲歌HK3*QwTGK#<7#?M#T n͊= |tsU,x;CfgXz;}ւryC̋,jѨǃ7PF|/ q+Ibbq6F(l;,Ĥ;DkD םӈg2eI$eK+t$ǑLŧT/UTM ;~qs լ0Oz%Z{< wi UDd X J V C AP? "Q2&"o*]1`HE=;8`!&"o*]1`HE=;  7YxڕToAfvQ[hvhhV o]Q6JEaTLcx2?xɚѣ/0Ƴ1doFl{߼y{o~0cpB|25<;a[.sFR /|lz ?bV$Z !"Ch/ˌ\AaيC-zcqUxV-n_qWfp^F9l&3kZ6fw|*D nFP>ዴ}3^3zg=6Ea7lYGYj`e]_7~XMxŧun*,a\>A}M.{USitWip?گ=]TklpC'U`Vy#VċDj%wU'Qkz] ϳzE7 őKnr)TΜ&`sV7RPݔVmN XXKj)K1( 飜)NuQ&4Hb)P\ގ"*Ay!3t"2pQp'atep^yLd9ZB'*>z?+\cAޞ2QJ|as+~?ӺDd  J W C AQ? "R2fowXQU;`!fowXQ xڵTMkSA=3m>KڂbP6UpY1_XJ[M1FlW$ŕ_Bt-"Eμ&V[Gf;w3 ݀6QEF<Iy8;ڎc~!a:y_S{^ljb:kI/5In%ϴwZob3t(`U`ڔ[J>aa uOɯ&^\X!:B]>W_Uz&L"˟xΜd].,iyߩ~us\> B!q45NUoybEv*k㹹*sQeg>-bJV3RFW-k/Uy,'r0Muv4YKFSvϜ&JZZ,CHǥ%̄USdLe%Q #͌ cs3MR0LjUApmc{PS%E "OUC@ߊ RG)*rhLhYK3>M94jvi;q\T&ѡi;_Pq?Dd P b X c $AR? ?3"`?S20b(XR >`!b(XRxڵUOAfv+R <-M- 4DOUmRm!%11&<`RJ '/&x`=oLq3{Yz@+[g̒VtmE1/ cB2/T&/ʃU5Q QE<.X^ Uc;XcL$Z.8d=6Z0}g@AIvۘ{b3+y0EѴZQmfېy2},"0{KV#Ks'V f-1~ublZ-|0acL;2dGݺ i[k£ bz&,x^'.N 80zscV}9<"膝!iڪ}_xdcu#=b49\ .>0>2q hR}j.y?37|.In|L>Բb>>fSÏ㭝@glx/'bc;gN|C%^9[)%HRb(+mJɘqaX[b 708= F$t,Q^^HQpyݜ/@.;F. k= iY7cSt AeBHC;pS0˜hWRִ6Lf,} O1iXߨ:C4Cx1kiS bٕ|y$/iDd `J Y C AS? "T2Kk{h Chr'B`!k{h ChrZ xڍRkQfۦi$kh,(m) &L &vՍ$L.VcDo=уyo7HۙyxCX0N֜l&)8")Ҫeqh ²Y\C.ah&dO3xePsKkE ]acEw^tR@Z0f{ޫ#&T{i妜8퀷FZ~_)0(l<#aٟHHϼH4kbmOh6뿹Jut~߿D<4W}qtv~i^9s6D͙ek{=eDkn.#us:l>pdmV8qYٲjczWTj>;!)XLR]jZyat)4Ͻ+$Nr1'WSp<$@{ \Ͻ${ ɖp Dd [lJ Z C AT? "U2lX @ͥHE`!@X @ͥ`xRpxڝS͊A ĸQ4aaU$&\M1jd#ArRYY<{ xDX=3YQjeDLA'!+REr/+BKJ }-39 B$7lyx3ρ2PGOgH rJ@ }"餽-7}?4[<`7;` w=0og]a2x٦5+qfuQkY0sF+jw-T$kc{YIբW8U7;&GhrCRQk.x9tmi$B\8+bY:@:!K5 F#K76+[mֿ^xr!7ٶȶJrUuΔ*w+{P~AN <TV$Z9qL Q-W= B_eōl#ߗnIH_N2 p^*9pni7ܠoDd `J [ C AU? "V2LNjbu; ~Cb\o(H`! Njbu; ~Cb\oZ xڍRNQJ$h%%l6%.ڤv.kxv.} >13q<ԟЛ|sι0 `CLS$qZbqhܒ{iD.ahƂ/dg {PqKjy ]CzeAy5IiͶyO:m8إzȘәg8j,vݦUFdv[c;QiӮVcWݾ%x9m#(wю#J g7iͶ׻ IIJf2 R @K8q,/&y]15}-r#8[W=j![z%ڰ{hhiXLDd lJ \ C AV? "W2^s2YE=K`!^s2YE= pPxڝSOkAM5 dXD z(M%mKJ0)cdDlPO`G{E9ˍ&k lWϾݭӫ\}}b8 lgZ«~̷dDd +J ] C AW? "X2O6}.wWN`!O6}.wW hL(+hxڍToQy4,=ZIV.?bmCGD* ]XX1Fhw.=ziū7CyZ7>{pnI*ǯE83e]fGchO!sAy|a7 qz{ہGb_~`(6_ߓԶUWL^ r4]{hkAjYg 7fTFf*wS3Y-D{#TݬXMBDdfղm"7ui%K9WSodϫ[+)VT\w{su2bq10Rs!U@ uVg7_qx$v(ލzI DHi]iϱM,LueG>=ebt_31zk;_6`1iVm J>P Dd J _ C AY? "Z2n_,Yp;0݉JU`!B_,Yp;0݉ x xڍSAkQf61mMkQ4Զ)*n`q[ $$L@Thx.xI xW yoS.~{IFcb"C9Β#B)AWOs(YR(j4ˤc<7ԢaXzdֽ'{)^TI8/mIH>+* *?БL'|zī<^)oW:V~sk9i~4˛g(.G R0sc7no=Ct|5ϷVAcWŠ],c\(xִ+`dye`Y s[5Spzm3ERWaES4Fb֠Ӫ -fƪ00QXUpUi9x+?̱4Xqx{ON<7(ʜ.)}ہSg}LHgřSEɦ7P.IwloD rzcv:~>*͇Vĥ`׶zjq*7{qLqPȘ׈ݮW.3^p5wYgd* Ft&yQ/ԋ.\ ]rX ~HڈM yrxE>ޘ!q9xp YIN!gK8Aid0ebA ~0pF?Ts^߱ΕSӶADd <J a C A[? "\2W+RͅCwM[`!wW+RͅCwM`` (+ExڕToQPJiviT$Ѵ4xt{!鑢lV7!>7G *0òN۱{UDd J b C A\? "]2j|nˬ"!hF_`!>|nˬ"!h ( xڍSMkQ=N6i$"V&BLIMɨFRmuQ.@dv}o* hy9#'&Hdj4$gS)\A~,|AuX4J\Bxb^OΆޑ x̖*qwGsR)p+hEjMEj{ΪF:~^TU\^Y`htiLø_:*Sٮ'z`n1YJ>f1}}<,ٵJwڏkWϧl`1n8TQa9Xu6άK|)^uJN9PJ< PbF7VU6-d"/'%w6dq-Q;ѹX8GGl}YzDd DJ c C A]? "^2&{!S6$b`!&{!S6  *~xڕTOkAo&&AMEc@KK&)i1h H"ŃR C#$Jݛf6 ;3}7f& DQgЌu])y)ΐF (Ģ4~.+ h$LZ[ԴT7 2!@3Ef b1WoQN j2yұӤLƤ0EӲnzOD6Vw,len%Y6m py>eGS᥋gUs- i!^I NOJn@@Os#>] !+2j?RdړH b0`mg!KF[qOÁnV= NMArX|uHTT,} K%,g^8! le`P!xu+$f6wK?LeDd J d C A^? "_2S7.Uхe`!S7.Uх $ixڍTkA~&٤-A,z($AA6 BJjR $1.6D$'&d[/9TP<'Y nKi&y;o;>& 7 MӔ$,#Ux bGȻ 2!X"GV|'6*zqDIC-b%`ܔ`䄠T֍kcz}VhNA6vS 0d7ko̢?O>%YbmVT`,Zxi%9Dd J e C A_? "`2/Cg_wi`!o/Cg_P =xڍSMka7i>ȦX44n`&EoIݒHNJE"OH/9^B-R|Dd lJ f C A`? "a2X7OA$z]Yn+r3 i_äM\1\ed9$]~F@+d~zbm$ʀR4h#"L #/ZeR11yM:〹h5m)yL~ Tirl3E?BPtreꢳv.ϊ7T!̾纠MZY޹W-ٻgI;ݜD_* `fZn0=º0O ,y#{;-߾lK_HK2!N3-^z5vlk]`w=~|Zx|5 gBa1[?rJUAYߐv bgUd\lC>+3+'Hѣp hf1) g\rr%RfL8AFa0N(<6?pm7ʒͤbz2۔ *6M5MZw@I aLD.{mڨ\J0d rm(%m[ev[@;P] nUl} Dd J g C Aa? "b2DNqhs9_ o`!Nqhs9_0  xڍRAOQfB)51}hM<\J[qMZJml/ M+83ʢo(s:kt#]Rl.Odé7AOoN?Hn&va' 8qo&ю~o42ݹz]@mA^jij x'߹ȆYIE'Lgg)>D=IIvw䱸Tԟ /n{ΫzȥڰS66" 7~^j<==F[ .j1*Gv42M^+Gw, bžSFlmqAfDe屷,+CPs3Oӝ`L7j^gݓDd [J h C Ab? "c2IQ!YěGR:2-!%r`!Q!YěGR:2-!` xRxڕSjQΙi`&Ŏ+2i&%dR\NF44oP|]}>soED:̝s]"`|!q dd1LIhMִFK1:fA 7C؃-3e9IN,^94y{":eQB}\LtP8"-vG^x $vM9qJSCޠY6im5`_3eEJdv_5{>0y=ë݄zft޿2.\@QvQIu4lly#oެ^PF0Qe,ڪOj[n3~S"ǷɎ/ȶJLOf[]_z]+GRlby$! eZudxɌ*/)KYוN#iԅ3*wXm s.s}/0큍zwTƣ(w~Dd J i C Ac? "d2VW)Tx)m2u`!*W)Tx)m p xڕAkQ36 즵HQ4T$EAoAh&fFIQ?BٓOUdo뼗MD.y2a0C= YLX(D[s%U5F23Ʃ QO8H}X(V ,oYԼQsZxiB߰)'#]ސ6 ܾ,w㊋Np-Tr Ҟp@4D OJJHϩ 3noǽuj3s!k1a3Mؚ,ư{]iEhx7h6%pQk^6Ng;~{M.v5&~ \?Qeۮyǵ?.ip㊲LGQ5,H{n}+.mfHI,4DҠJɆjON/KmYS@|N.,5/Yy ]m`: )%Awgm c[Dd J j C Ad? "e2s,ɿn{P(Qux`!s,ɿn{P(Q2 @axڭkQ6YLCQPS)d%54 CFaUԂxK`{ ͳW=Yl>Jl|vfa09=F72Z}:2th=D)?\UGq`@>KlzPh?2Ex!`X31.IfQY955GO `) ZX5.5jS_zë,0iux(?ߗ!p=oQqKn<eWXtndQَS /R>ܨV9l-t6||\.=ZK`kMY|vQ GkH >O8=l?a],>d%g +t~Tj5YnUu3Wo 虬;ȝY1A̖UӲ`/GT4[D륋$`+v7Y")7Mۋݕ82$$KR_gTCcA6+$GʐKA!| y,^~/OZ }~^D2&?)LI)B}.$;*+w_)wL@hR3 ƋBV^KP~>Zov {o(M .*ܒ˓S᢬1&l;8q}QԀIQ\.*\ &$<_\iʩ16]5gû́gI;)z!ϼ}jl .5VnsQ[*s{ -~ֵ d" Yg9oFW fKtW*z:WnV돍|ф6y۬ZQ1a/7i8^v3z +I`ekVwDR(K7kMrVWjȒO2%=gCT3F Ҙ=1ɑ"hNZ}I~LX t# Ȝ$󇝄K by߸8SU`Կ=.Jpo`gDݞS3R#}@bi57ۛ\Dd  J l C Af? "g2/]Vs&eCompObjor iObjInfo OlePres000qs Equation Native  (                     ! " # $ % & ( ) * - 0 1 2 3 4 5 6 7 8 9 : ; < = > @ A B C F I J K L M N O P Q R S T U V W X Y [ \ ] ^ _ b e f g h i j k l m n o p q r s t u w x y z { ~ .1  `&N 5 & MathType "-@7@F "Times New Roman- 2 .. >"Times New Roman- 2 ,t4Times New Roman-2 n_Times New Roman- 2 tB."Times New Roman- 2 C2"Times New Roman- 2 tt ` "Times New Roman- 2 ttKt^/i"Times New Roman- 2 tqTimes New Roman- 2 HyTimes New Roman- 2 Q\Times New Roman- 2 KqISymbol- 2 2 =-+=-,k 2 &MathTypeUU DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    .  t =="- "H C t  "K t ++ t !  .  t == t "-Q K,t  t &  "-"System- ~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    .  t =="- "H C t  "K t ++ t !  .  t == t "-Q K,t  t_1233261765vFP%RP%ROle  CompObjux iObjInfo  FMathType 5.0 Equation MathType EFEquation.DSMT49qaOx <|.1  @&N & MathTypeP "Times New Roman- 2 OlePres000wy Equation Native ' _1233261766t|FP%RP%ROle + .`.Times New Roman- 2 |&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    . & "System-~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    . FMathType 5.0 Equation MathType EFEquation.DSMT49qO CompObj{~, iObjInfo. OlePres000}/ Equation Native ? 8.1  &N & MathTypeP 3"Times New Roman- 2 ."Times New Roman- 2 .`4Times New Roman- 2 ()&`Times New Roman- 2 UCCLJSymbol- 2 0` 2 =&(MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    . ==U(C) C . & "System-~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    . ==U(C) C . FMathType 5.0 Equation MathType EFEquation.DSMT49q_1233261767FP%RP%ROle D CompObjE iObjInfoG OlePres000H \Equation Native Z H_1233261768zFP%RP%ROle ` p4 .1    &N  & MathType` 3"Times New Roman- 2 .`Times New Roman- 2 ()6Times New Roman- 2   "Times New Roman- 2 K`Times New Roman- 2 UCCQLJZSymbol- 2 0` 2 =-&5MathTypeUU)DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  U(C) C . =="-Q K  & "System-,~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  U(C) C . =="-Q K  FMathType 5.0 Equation MathType EFEquation.DSMT49q2p: .1    &N  & MathType` 3"Times New Roman- 2 CompObja iObjInfoc OlePres000d bEquation Native v L.`"Times New Roman- 2 ()Times New Roman- 2 ()~J "Times New Roman- 2 K`.Times New Roman- 2 CUCQ&LJSymbol- 2 0` 2 =-&:MathTypeUU.DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   C . U(C)==("-Q K ) & "System-0~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   C . U(C)==("-Q K ) FMathType 5.0 Equation MathType EFEquation.DSMT49q_1233261769FP%RP%ROle | CompObj} iObjInfo OlePres000 Equation Native  (_1233261770FP%RP%ROle   { .1  @@&N & MathType` Times New Roman- 2 ()Times New Roman- 2 -"Times New Roman- 2 ttsTimes New Roman- 2 UCjZSymbol- 2 | 2 =&MathTypeUU DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  U(C t )== t & "System- ~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  U(C t )== t FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObj iObjInfo OlePres000 Equation Native  `p\ ..1   &N@ & MathType` 3"Times New Roman- 2 .`"Times New Roman-2 ()()(~~~Times New Roman- 2 ) "Times New Roman- 2 K`.Times New Roman-2 CUCUCQ&L JSymbol- 2 0` 2 =-<&MMathTypeUUADSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   C . U(C)==U(C)("-Q K ) & "System-D~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   C . U(C)==U(C)("-Q K ) FMathType 5.0 Equation MathType EFEquation.DSMT49q=  .1  &    $ !&"#%;'(),*+-1./0326457=89:<P>?@CABDHEFGIJKOLMNQUiRSTWVYXZ][\_^b`acfdeghjnklmotpqrsvuzwxy{|}~_1233261771FP%RP%ROle  CompObj iObjInfo OlePres000 2Equation Native  _1233261772nFP%RP%ROle  Nx & MathType "-@h !"Times New Roman- 2 .MTimes New Roman- 2 ()(~~Times New Roman- 2 )N@Times New Roman- 2 ()>"Times New Roman- 2 KMTimes New Roman- 2 UCQ NLTimes New Roman- 2 CUC&LlGTimes New Roman- 2 CC78Symbol- 2 8 2 0M( 2 -@ 2 =&oMathTypeUUcDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    C . U(C)C== U(C)("-Q K )C &  "-"System-d~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    C . U(C)C== U(C)("-Q K )C FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObj iObjInfo OlePres000 Equation Native   .1  @&N & MathType "-@r  "--   !"Times New Roman- 2 .NTimes New Roman- 2 1@ Times New Roman- 2 (~Times New Roman- 2 )|Times New Roman- 2 ()Times New Roman- 2 ()p "Times New Roman- 2 KNLTimes New Roman- 2 C@b Times New Roman- 2 QlLTimes New Roman- 2 C|Times New Roman- 2 UCCLhTimes New Roman- 2 UC@Symbol- 2 =- f  2 0r( 2 ( 2 x 2 ( 2 &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    C . C== 1 U(C)CU(C)[]("-Q K ) &  "-"System-А~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    C . C== 1 U(C)CU(C)[](     "#$%&'()*+,-./01235678;>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ\]^_`adghijklmnopqrstuvwxyz{|}~"-Q K ) FMathType 5.0 Equation MathType EFEquation.DSMT49q F B.1   &N`  & MathType "_1233261773FP%RP%ROle CompObjiObjInfoOlePres000Equation Native 0_1233261775FP%RP%ROle -`  nTimes New Roman- 2 /`:Times New Roman- 2 !Times New Roman- 2 /nZTimes New Roman- 2 MUMU>4>Times New Roman- 2 CCnSymbol- 2 `T 2 =-, 2 & MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  =="- "MU/MU"C/C &  "-"System-~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  =="- "MU/MU"C/C FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObjiObjInfo OlePres000!Equation Native 44  E.1   &N & MathType "- nPTimes New Roman- 2 ()`:Times New Roman- 2 Times New Roman- 2 ()nTimes New Roman- 2 UCCL$Times New Roman- 2 UCXSymbol- 2 0`T 2 =-,vh 2 &!MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  =="- U(C)CU(C) &  "-"System-~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  =="- U(C)CU(C) FMathType 5.0 Equation MathType EFEquation.DSMT49qD .1  &N & MathType "_1233261776FP%RP%ROle 9CompObj:iObjInfo<OlePres000=lEquation Native [_1233261777FP%RP%ROle b-@r  !"Times New Roman- 2 .MTimes New Roman- 2 N'Times New Roman- 2 1@8Times New Roman- 2 (~Times New Roman- 2 )l*Times New Roman- 2 .  "Times New Roman- 2 K"Times New Roman- 2 KM Times New Roman- 2 QNLTimes New Roman- 2 C@Times New Roman- 2 QlLTimes New Roman- 2 CM Symbol- 2 -@ 2 =--=,&MathTypeUUwDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    C . C=="- 1("-Q K )== Q K "- &  "-"System-x~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    C . C=="- 1("-Q K )== Q K "- FMathType 5.0 Equation MathType EFEquation.DSMT49q " .1   &CompObjciObjInfoeOlePres000fJEquation Native tN ` & MathType  "Times New Roman- 2 >"Times New Roman- 2 0Times New Roman- 2 ()e "Times New Roman- 2 t "Times New Roman- 2 t_"Times New Roman- 2 t>&"Times New Roman- 2 tTimes New Roman-2 WUCdtb"Symbol- 2 = 2 ->| 2 =Symbol- 2 =*~Symbol- 2 &dMathTypeUUXDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==qU(C t )e "-t dt t==0t==" +" & "System-X~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  W==qU(C t )e "-t dt t==0t==" +" FMathType 5.0 Equation MathTy_1233261778FP%RP%ROle CompObjiObjInfope EFEquation.DSMT49q3 p, .1   @ &N  & MathType` 3"Times New Roman- 2 .`7Times New Roman- 2 ()w&"OlePres000TEquation Native X_1233261779FP%RP%ROle Times New Roman- 2 t"Times New Roman- 2 tt.`@Times New Roman- 2 SGSR`Symbol- 2 =-&EMathTypeUU9DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S .  t ==G(S t )"-R t & "System-<~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S .  t ==G(S t )"-R t FMathType 5.0 Equation MathType EFEquation.DSMT49qp d.1    &CompObjiObjInfoOlePres000Equation Native N & MathType` "Times New Roman- 2 .`Times New Roman-2 (, )(, )tqt"Times New Roman- 2 t""Times New Roman-2 ttttt>:,`FTimes New Roman-2 KQKRCRStB>`?Symbol- 2 =--GX&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K .  t ==Q(K t , R t )"-C t "-G(R t , S t ) & "System-Ь~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   K .  t ==Q(K t , R t )"-C t "-G(R t , S t ) FMathType 5.0 Equation MathType EFEquation.DSMT49qp m.1   &_1233261780FP%RP%ROle CompObjiObjInfoOlePres000Equation Native  _1233261781FP%RP%ROle N & MathType` 3"Times New Roman- 2 ."Times New Roman- 2 t`@Times New Roman- 2 S&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S t. & "System-~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   S t. FMathType 5.0 Equation MathType EFEquation.DSMT49q5 y.  &N@, & MathType zTimes New Roman؟ww w f-CompObjiObjInfoOlePres000Equation Native 2 ()(())({, }.=8%Times New Roman؟ww w f-2 (, ))~q"Times New Roman؟ww w f-2 Ctttttttp.=>)t"Times New Roman؟ww w f- 2 tttR,zFTimes New Roman؟ww w f-2 HUCPGSRQKRUX tyTimes New Roman؟ww w f- 2 CRSSB>z9Symbol Th؟ww w f- 2 wozSymbol h؟ww w f- 2 =+-+t. 2 --G+y&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  H C ==U(C t )++P t (G(S t )"-R t )++w t (Q{K t , R t } "-C t  "- G(R t , S t )) & "System f !-K xDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  H C ==U(C t )++P t   !"#$%(+,-./0123456789:;<=>?ABCDEHKLMNOPQRSTUVWXYZ[\]^_`bcdefgjmnopqrstuvwxyz{|}~ (G(S t )"-R t )++w t (Q{K t , R t } "-C t  "- G(R t , S t ))_1233261782FP%RP%ROle CompObjiObjInfo  FMathType 5.0 Equation MathType EFEquation.DSMT49qg F v.1  @ &N  & MathType "-@ `Times New Roman- 2 0v7OlePres000 Equation Native  H_1233261783FP%RP%ROle &"Times New Roman- 2 C"Times New Roman- 2 Cj Times New Roman- 2 H`Times New Roman- 2 U{Times New Roman- 2 CjRSymbol- 2 `w 2 =-==^ 2 &6MathTypeUU*DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H C "C==U C "-==0 &  "-"System-,~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H C "C==U C "-==0 FMathType 5.0 Equation MathTyCompObj'iObjInfo)OlePres000*REquation Native @tpe EFEquation.DSMT49qF* .1  &N & MathType "-@ `Times New Roman- 2 0 .7"Times New Roman- 2 C "Times New Roman- 2 RRj Times New Roman- 2 H`Times New Roman- 2 PQBTimes New Roman- 2 RjRSymbol- 2 `w2 =-+-G=,   2 &dMathTypeUUXDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H C "R=="-P++Q R "-G R ==0 &  "-"System-X~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "H C "R=="-P++Q R "-G R ==0 FMathType 5.0 Equation MathTy_1233261784FP%RP%ROle FCompObjGiObjInfoIpe EFEquation.DSMT49qFz .1  `&N  & MathType "- ?"Times New Roman- 2 .. `HTimes New Roman- 2 OlePres000JEquation Native a_1233261785FP%RP%ROle hr "Times New Roman- 2 SsnTimes New Roman- 2 H`FTimes New Roman-2 PPPPPGH*RTimes New Roman- 2 SnSymbol- 2 `2 =-+=-+G,~H 2 &MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P . =="- "H"S++P! P . ==P"-PG S ++G s &  "-"System-Д~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P . =="- "H"S++P! P . ==P"-PG S ++G s FMathType 5.0 Equation MathType EFEquation.DSMT49qFN .1  &N` & MathType "- {"CompObjiiObjInfokOlePres000lvEquation Native Times New Roman- 2 .. `4Times New Roman-2 <p "Times New Roman- 2 KnTimes New Roman- 2 H`Times New Roman- 2 QTimes New Roman- 2 KnSymbol- 2 `2 =-+=-,~< 2 &}MathTypeUUqDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    . =="- "H"K++!  . =="-Q K  &  "-"System-t~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A    . =="- "H"K++!  . =="-Q K  FMathType 5.0 Equation MathType EFEquation.DSMT49qt :z.1  &N@F & MathType  3_1233261786FP%RP%ROle CompObjiObjInfoOlePres000Equation Native _1233261787FP%RP%ROle "Times New Roman- 2 .`FTimes New Roman- 2 Pz&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P . & "System-~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P . FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObjiObjInfoOlePres000Equation Native t   _."System_0^ _- @Times New Roman-^,--.  @Times New Roman-. P2Times New Roman- 2 P2.tTimes New Roman- 2 P"System------'"Systemm|^^ j^m0-''' 2 d^ --~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   P . FMathType 5.0 Equation MathType EFEquation.DSMT49q {j _1233261788FP%RP%ROle CompObjiObjInfoOlePres000Equation Native 8_1233261789FP%RP%ROle .1  @&N & MathType`  Times New Roman- 2 (, )a"Times New Roman- 2 tt>Times New Roman- 2 tSR~,4Symbol- 2 G=G&&(MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Gt==G(S t , R t ) & "System-~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  Gt==G(S t , R t ) FMathType 5.0 Equation MathType EFEquation.DSMT49q {  CompObjiObjInfoOlePres000HEquation Native .1  @`&N  & MathType` "Times New Roman- 2 1Times New Roman- 2 ()/0&"Times New Roman- 2 ttttTimes New Roman- 2 RS4Symbol-2 G=GG=& GJ&wMathTypeUUkDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  G t ==G 1 (R t )!"G t /"S t ==0 & "System-l~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  G t ==G 1 (R t )!"G t /"S t ==0_1233261790 FP%RP%ROle CompObjiObjInfo FMathType 5.0 Equation MathType EFEquation.DSMT49qt{.  .1  @&N` & MathType` "Times New Roman-2 OlePres000 VEquation Native _1233261792 FP%RP%ROle      !"#$%&')*+,-./036789:;<=>?@ABCDEFGHIJKLMNOQRSTUVY\]^_`abcdefghijklmnopqrsuvwxyz|}~21212S@Times New Roman- 2 (, )Times New Roman-2 (0, 0)x`&"Times New Roman-2 ttttt>z:Times New Roman- 2 SRRS,z4Symbol-2 G=G=+><&u & MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  G t ==G 2 (S t , R t )== 1 R t ++ 2 S t  ( 1 >>0,  2 <<0) & "System-~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  G t ==G 2 (S t , R t )== 1 R t ++ 2 S t  ( 1 >>0,  2 <<0) FMathType 5.0 Equation MathType EFEquation.DSMT49qt CompObj  iObjInfoOlePres000 "Equation Native (<.1  &N`5 & MathTypep "Times New Roman- 2 0.5o7 "Times New Roman-2 31212LNTimes New Roman- 2 (, )Times New Roman-2 (0, 0)x`=_Times New Roman- 2 tB&"Times New Roman- 2 tttt>z,Times New Roman- 2 SRRS,z4Symbol-2 G=G=+><&g&)MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  G t ==G 3 (S t , R t )== 1 R t ++ 2 S  t 0.5  ( 1 >>0,  2 <<0) & "System- ~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  G t ==G 3 (S t , R t )== 1 R t ++ 2 S  t 0.5  ( 1 >>0,  2 <<0) FMathType 5.0 Equation MathType EFEquation.DSMT49qr 9.1  @&N@ & MathType  "_1233261658FP%RP%ROle 1CompObj2iObjInfo4OlePres0005Equation Native P_1233261672FP%RP%ROle W-9 ' "-M  sK Times New Roman- 2 d(/)9GtTimes New Roman- 2 d(/)8Times New Roman- 2 //a "Times New Roman- 2 RK] "Times New Roman- 2 RK]s Times New Roman- 2 PP]t<Times New Roman- 2 KRTimes New Roman- 2 KRPPU]`Symbol- 2 s`Symbol- 2 =&MathTypeUUDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   s== d(K/R)K/R d(P R /P K )P R /P K &  "-"System-pД~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   s== d(K/R)K/R d(P R /P K )P R /P KCompObjXiObjInfoZOlePres000[Equation Native t FMathType 5.0 Equation MathType EFEquation.DSMT49q] .1  &N` & MathType o"Times New Roman- 2 T "Times New Roman- 2 TT>["Times New Roman- 2 T0Times New Roman- 2 ()dT8"Times New Roman- 2 tS"Times New Roman- 2 toFTimes New Roman- 2 KKQC?"Symbol- 2 => 2 =Symbol- 2 =+-#N*Symbol- 2 &MathTypeUU{DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  K t ==K o ++q(Q T "-C TT==0T==t +" )dT & "System-p|~l~DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMarlett!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  K t ==K o ++q(Q T "-C TT==0T==t +" )dTOh+'0h  $ 0 <HPX`PART IVARTi2i2i2i Normal.dotPermand2rm.[cmD`!/]Vs&e[cmDxڭAkA߼$Mݤ(iLB/Biͥ%6-CEI&+҃XzcEC{z~F`|3Ijipyf7}s F72Z-g:2z km#d]pZbBaFlіeRŨF檕_ P#Q|[Lm۰y(nWիW';V̑5&LjMowQjS3b'΁isp܈&v֡.3}msԶS.UxrԮ{w9j^f Z@AͯTW͎c—p=&NG$ߴ^FӮ ߙSAv Ng'J+-b ]}h䊏:=ч8eAuWb兡hD >h 1 UEH$]oR=[6~blb㙜V kT%硾cQ|/5fq~xh5[f. ܥp$O E6P+^j\סs?-?)ɟ3&Ga~YaͲ~?j 으^v\r8驟 InH=J]? ۨӌGNz\øJO㚔즢֞cܽ| WqbԹEB][P7{+5ߋeYEPThW 0xUۙh6;.dC~Wu~^wOg2ڌPgn艗{#껑\8Or/ FܸhR$Jq30JL?$D;S!ި( P%|4@,t|p8KZ(_:_,D Bߖ`MJed}xRHB0}tAMVتѨGv_=g޻}g] o2 ʽokլk Xk9pyFi^ 'Dd J n C Ah? "i2H6YM5f*rs8F`!H6YM5f*rs8x@hxmRMKQ=N0c&DP ] jcH*-5BRho/hӯpoi1PQ=fݏ޹ Lڎ=5? [A)OR,(&qB!--&u,“E؇Gre\(؝reZDst'ncN O] BQ֝v|߹4vÿY fdLNb$lG?cلX[=_˗J nJ6=6Z-c'SϜ6"ߧER~ܬq(ۼlU "BUs:"s)eߖtn]uguk*OsYyW~,C/V/X%uF8oQfaT]x PS~'Sj7RHSθ>iwd Dd J o C Ai? "j2nܨw! _@J܉`!Bܨw! _@Pg xڕSkQf6i$R4,iԟǀM)$xL.HI"R'г@ś7O"{u&̼o޼G/0 &BP,yړs%U(G18EX)c2CY1䫝gs(hi t!4+{Zޒ6 Tk82٨nJ\*JSiр _ºη|Xĺ?8`%jZB=r$7e}8ǟ nVnQT1&\Xܡ.hryr8[̗n3kVRm'Ni#.aAUsZ*竭 —sUp݊Uvٸ7^q+AfVs*n_{}*Zd+9W}VxڍSMkQ=6IkFc@i IN'bMIM&:h i$HV\W."]DPMIB$;{8} O @`D%t$At ޕ{A${B8J tk5|@Dzo?f)>cL[D{Zސ6FR%YNV^+9W~Se+䲭9,iq!8{Glmlu$;;t~]hth{_Uq*vsUE |[|ށ[!{]_ӭ[d PTr0ybVfUhgˏ& ò*OrjZt'5kf#thf?TȄv.:"{FRvf NGGn}9SLӎ !D ?E[/`kӼ[ʋ1DA:>% ~yDR'78C9팂EbB`Snm}Yh4nY/7Dd  J q C Ak? "l2V,4 1RY9u"`!mV,4 1RY9: P;xڍSMka~fmIZD`iMIm$ꢁ$"mԃGŸ`e/4λ mγ3><3;C 0PטLX^g{t\aB~Ѿz8:I{rf-ۏMdL(>V_=[^m|f6#+ȝz*NH\*UJSIV氽?.J`,M4mg@F89`S_<ÞÕӱ$1њ:Ky9t#ڵ{*]v/V7T+麓sS($#l:0uW fQ`&ž Ú*1Osܨ&1~i>I9+Us?/^<aeP t2\q؞b/ bR= rIA$m`§鈼0 W7Ri1'EP<2cF6>k✌&$iU8*.ubYFI;}jG%o mk]k7[fQdfeW7#d .#}DZ}^wÆ GDd  J s C Am? "n2Ǧ~HVYs"~3Q`!}Ǧ~HVYs"~3\ KxڍSoQ<[٥cHҦ4#$ZG@( å6dk9xj@k8-]7~3;#08FBUP'qqY$ Mz(HamN1AW=5X|G%OLȔsP,㊏@'L3CB/+W:*V-mϙV@8VضM'ǙV6{bKa ƻ?P *葄~燳g{ ǪQ+[ӾZkgK^7PW~wlkadt:FHJ!z}>mapRj#L=*oڕrQ~=ugN!c8א) z٩+'B)+NI]_LSvֵ{;UӦ 7κ#Lv +n[,zPHҮfP'$.yCM(l$v\}?{wFQmz7_%MFbY\Q*lQ|)𾩅szw'lM:-R홸Dd  J t C An? "o2?\_Ǽ!x`!?\_Ǽ!x ]xڝTMOQ=NJi)Hhi"iI jbdiH  ˦De +Wla㖕q  fVX/J`;yy !A^!I< jvdT\#IDE8pAC(@1|ksuYQե'&sd zv=gZ8 elhUbb4zdEd' 0A#F8З*5/s{u܉H$P8jyzv^W5ǯUf$SIMlŽEBW-N~v(4L O_SOޠg|:h,`S'ڧa1Gv54M;al={4{$u7jU?{OkSp[~6Ohɴ(`nsGέ o+B僂Kwy'.WֲdQY1cHLq{Fz\I(_^&˵ƲMn\g{z {d*c(YH VEp/~q]7EQ5Nx w#ʃM)8We٥Ah+" KFyy&%R-Dd lJ v C Ap? "q2[ 2OP-Tac#k-`!c[ 2OP-Tac# Xyp1xڝSkA~ͮk$E4b ^7`/[Il.I"? =RՃo^Ifv;͛7|EQ$ v@6BGd,\!^!Jgk]<0`L|dV+Ns6z챏ߘ/eՙ9 赐(ͨaH(O`V}?7YkuAf{4a|9OyՐn9Yt`;m/d#5~c1`j]0+5z2bcf#bWe|.HVG{6?s[5BL;a fpZ2\T:\gwZ].6vWi>Gޔn[^_.Bf Wq)oU|3y˵*1^IZ 22erg 5}YrH8XP)7/ Ҷ P~)]ۦ>.[u^!zѢ%~M kltVGיֈjD(^{Dd @> w # Aq"r2%^ŒZ`!%^ŒX xeQJP=sk}@RkA (jUpo""Q] ܛP s2wsfa A@ҦIc٢\PTG11"E%sf+b(YKX )#KD]nR9M*۷vjo1+nD: 6lAy>q=3Wz AWK= q촌F#k\®Xvdj@`Wۅ7.Z(tզ`!l\a>@`Wۅ7.Z(  d:xڝoA{[Iv)r0i5m#VM]=HCucIa56DSO7I5:3}oͼAhi"<"ce'!R'TA$܂K0` L|`7ςWP+.3/"ʨof B$AD:x>Jv[v r~1/i+ϻCWaGjhȕ:#JJGD1ſpC.8 7h3ڿc/ф{?i{A[7F0= zsaŶ>{&#^2NqRsS\\ZUoYF[gﻥ }>"NvQvTunY^/ V3U))UVaaX֪PYtVW [F9 $n,PC~V"93Y!oqׁpkKꫩ_6fE ᝝LxɚpL߫9nNDX'B.Dd  J y C As? "t2r LS\y `!r LS\yD8xڵUߋQ>kQ["rjWYLqӥGJPcٿAؓA=lT;\-9w{C=xx)!Hh3rBD)'C &R bѹ#>{ bxD RNyW2 !}Nhu@T#V[{ro\L%5F'CDO1rEFZRZ7}gßlۣضM6/X{kGEJX6k옔u@bF;~e]'S:w99)8G=3ZWz*L:Nkb3^ڝefFUg?Y{U$)v:|"Axd>Sg򺬪˥zfzKɕn+u9WLe>GzVVjl|4SJ~p/Z%9•z49A-ZED`g d1,x .PVpf|x-Jf9@1)<k;ΨW$#e,i4-l6B & &G_ }d3z~OACBfT~~3fo|Dd lJ z C At? "u2 oJ[;z`! oJ[;z" Xh(+xڭToAffi٥&ƦV&i@셆oHe$P6޼C/=?Ã1doFj}o{<7 8|8cpdZep4o)̨XDWN4I_ < C"oju٨E(*m=VteglϿe|ٳ纩1Z鹪}TD 'Ɣ@v6~ָ\!U/VJ`l=}\yGs 1 igs*g/ڿh}عֻ}6cD60-EJ|0Z)#k.~wE+ϪE*DZ s8wu=ތGE+K/qʑQ)q36WۤDX.n78=3m0dFa),K~oV,AA& _[>2\ DƏagVo `ؤ4G8Oz.+{ݖfщտ)@^3yV2[Dd J { C Au? "v2v8^rR5`!J8^r,@ WxڕSkQf6MɦX4,m!)*"&KzI IEIS%'!+GOzO^/Qٛ`6.;ߛ7{߼7CEPOH>&BX,h4Ҟ ]֞LWh1F1NȒsq`ax ~Ʋ{Y0T.P, ]teA85ҢpD0j;Ai7Ɗ2/[VT N(`PvlgNX#=UVITҔZo /z^y 9L՞*St* gQ/dΞ1+v~i5q<Ncww=jGn!!aMݦUHpmE_ߴMb*vUr܂o7VM+3ڶ?ԞgZf1$G"IԀN6Х 56jIOЋ,2.s,MTZo+21G'Ex<={R_Dd  J | C Av? "w2}\! 1I`!}\! 1  cxڕTkQݴiM*R׀-I!͇)4I[dErFA(<ś= Oٛ`%}d2f>)i DblMl5B4A_&iK \8+~|bL|Th=6J6ʔ#!o0$Sz30'IFoD|v^]hp>w ] D(҃DDA@HuTf^vw?s?, `PBf5Z*OiL׳Tnj$tP0 &zo_"+MCiy4Ql(^,¯R3?Ds$̤{8St ||j1G$ 4 #z<^g 3[E21#qF·z3.ooVp|zHVI\5Tyl:o-OK2" )+ OL/ E`,-5Y{g(I=^taB/r6kbtݵkIZ,8Xݹcf}3G-#t>f:yr|kƁ@WfG鬈;BL?'fB0P<@X0KOf 3$8MkZNSp{"$, ̭]]haџgZ2Pi9!1T:Rʼ"n@#՟_nC2jT5pPb@BQkCV&=n'I8QW6ɚzt}#ab6 al&{^ܙjDPM 4Xߑܰ3ӌl m%(E4iX.,*&Gh"Z0P&-Oh)5E/Mʙ<̉s'Vxo.~޹ynBDd lJ  C Ay? "z2~VGە_N`!x~VGە_N@ ppFxڭTOkQfnch]C S M0$xsuє)IU Ń?zN~`6 ogۇ @K!i<+U<4_%'ҌJyMk2+R($>2|ib B<[` ^ _Qk⚷i7X0#,Wtu?ߖ"7)!G`GJ~ax+80€'t,&&q6a6hW;)q#ՉX7q]MLIV#<{G$#+9ȉ;;w}0NuDTkyro7Vkks:J36)&11&ݖ;3> xK_,EeIw*L_QV'"0  V@g!>_NW Σr>2μ(ޙ@2TAvrWhrzv^ ^5܊=g5-6dn7Kp^[@Ox\fl1Q'LRW l' }tsTR B>U&oY'nb+:J[ՔDKuX:xG26W2O7pZ+S(.@a?*ʔ wk[fk X}hoaF |Ҫ-4e͢Gre3s*HZN3_;=eYi9_T0rHcLAw\6\;Ջl f @1jR[ENhQ+ׇ(>7Ҍ⋻gL|Mmocȇx=/ {{ߡi8K7?UrBi'ʴxǚ]أH ^eU-V;hG~|f 1Q8edO=g[Xnuj ǡ'<ԳNk~i5bF/jgC O֖CXc;wS j=PsjbՊ/D=2Q|aŽL^Y_V-3qm(/:%Xটw f63ӏX*{ ?~/5 }|+wZlciM*5۪\Yaw LFFBH4',?wcS6|x9xa<Ν\Fbt5VgnKz] zYo6)S3N|&1T|4$( љ˓zc$|Ȟ O C0cA4 LPqGI&8ꠁ|$H(zxdI<+;]XLS_ x亮$Ĭ*"x(Lt#V5PΑOv.yUaY{nq|$4m hU½Pu/ؗfjx/^<+FPbvH7M(ݺ4hT^Q/oуjèRVznEn#}[W֏5NݒID(ojv)L.&L(:gŏy1̀u^dkizYg9=ac`3‚ebce%p"C2GϷ ձtÀtL]_LT;v}4a/ҾuTDd J  C A}? "2yH}=yusf`!yH}=yusftRxڅRKJA}]1j>0cA4 LPqGI&8ꠁ|$H(zxdI<+;]XLS_ x亮$Ĭ*"x(Lt#V5PΑOv.yUaY{nq|$4m hU½Pu/ؗfjx/^<+FPbvH7M(ݺ4hT^Q/oуjèRVznEn#}[W֏5NݒID(ojv)L.&L(:gŏy1̀u^dkizYg9=ac`3‚ebce%p"C2GϷ ձtÀtL]_LT;v}4a/ҾuTDd hJ  C A~? "2WbTiRltK3`!+bTiRltKj@ |xڅSAkQf6iٍX%(zt)㪋$+F^#Bop[)>V)Z?($mv]Qy ?#Z&ɡP̯}xG0_3+QJ'$,] |LeUK'T"5cȉ1]u+nwԻfX+V Gq ~ooLj `Vgvanox!rl9#v,:@m"uogYT6Fh[~qY v[mTN8ў-Ov8JױyXdNp篌0RxVCWB/HtFF|Wg qZ V0{ް~$,Dd < hJ  C A? "2tξj L|;j`!btξj L|; `@Hb|0xڍSAkAƶi$cQQ4 Xڂ=M!%$)ӨF1[D{(E 7M [2y!L!CȡD%1 6]Q J4-ࡐXFܹ0a a {>13Ƌo|tڋ !Wqm`CLI>!E~i3~@RF(Uv+fo zeS|9 0> :JuqǬ3XJ?:9kjqűKv:]^n~bCwU}|w"$o]بZcIV[ӏ#,q_@;*E ?7s{@AuӬV*t㱝<[D4OI$nhVDD*͚8DIy[e3u3anֽVʌpZ^],=YfrXΔq@uA_, WymیN )nEw^.0=xԿ@7/E*K:oLaë WS-ǮMz4Dd DhJ  C A? "2 e>8&$C"[`! e>8&$C"[.@ *|xڍTAOAff `ڂV *mChBFŋVŃ &<Ђj7"+jhxUʅE٬1zw~)mCeȫ쥅L)t{^悹ㆹf{p)͡f4jFg{ H}GVg9a'J5AKGGGgFWI6UvMՆC/.J ɺI=&FcxKJ,w|?:65Y][TQT|R,.KBx{rǘ-5z9#虨,/%9 / .fY3ӳ/O̘xB92ʓd>c_@F*0J\@%cMDr$uz^Bҗskq>fG/#qiSm/,{i$Q{h1ٱaqtxa_@U2@_֌5-Toۖ*Dd h>  # A"2Y3m$\t:`!lY3m$\@ @h|:xڕoQƿy[jY=$zhT a ƄGDݴ$P 17L zbИ=x^M7q]Jnvoggoxo A@[!1ŗ B#`2\ĿR xJ)X]%L8f*Y<Q!Ο,رsp[I =;63$C\vd!&Z찫F%ҭr) f:%D{ۖ!MvB:a~dvz.t,,%|p{|G/]Œ([)+r~^)F_I&Gϛ%g.^ ۛ/I>\j)}07 uZMMR|^8 bQF \y%CWq.6qz@2AX-GZΎJuCv#5ƶ琯iưb8aRM rr@aewlrY>ր=^ӗ]hjh 7yDd  l>  # A"2 Ԟ>) 9A'`!Ԟ>) 9ApxڭT[kAl&i >6B6C mH[i铤.HiV$OEP?7 A( DIxfv$J ɞÙ3|͙#4>AIlugTdNp}`tw-sE^N/rfoƫܬ(9'gu*3&rwrfD&/1S(iQr)/ Me(*vK b;aE uQ<;˿A#LAYIQ5kz VXM o\{_wvı<\e3 鍻MJc3>D6;)%͕E7[--9^Ѭj }fёOt޴As_hYKh";53̯p,]3gVя\}p`.IDh*T4Y{Dݷ3 ; 7!;RRkxNE(# MfՕ )hAq|d)' z Bxؗ7"PZ9>,}& DDǮ!`@C"'+BD'\dm/@|Rsp^+G\߈꜕G/0As§6+K? ߡo[=t6]އ5wK1ni.nn!BsA_q~s]|Iʡw{\!E+N{U ezRjI7jwU^6SpzWn$v_^~!ޘ4_{Ne-SNʽzٱKl$D0k"4S^qe#$(_Z:Sls^ܜ+f/23e^s*nȪƊ^SE~2"3V-f=yeA@ߌHQba~!'ᒱ5 Bne~<TTF]a=G&xՖki~C9 d%^ڈÀo>pjzu&TnDd  J  C A? "2MASZ05JV`!MASZ05Jj(+rxڝToQ ҏcS+FB &6< FIb5=W /yowXlRݙ73y;p C  %~_Xr|smiוPa#ȃ.ih։ Cd>ך/t9" _43Ujzc5^}`Ja&!ug kܷ;NGLM:L6'LO!swف=)j= HXzlzmg" Q1TableGSummaryInformation({DocumentSummaryInformation84CompObjjMicrosoft Word 10.0@@R@RD?j՜.+,0 hp   Strathclyde Universityɩ: PART IV Title  FMicrosoft Word Document MSWoe8@8 rNormal_HmH sH tH Z@Z r Heading 1$<@&5CJ KH OJQJ\^JaJ DA@D rDefault Paragraph FontRi@R  Table Normal4 l4a (k@(No List :O: rTextd`CJaJD"@D rCaption 5OJQJ\^J@O@ rExtractd]^BO"B r Table sourcedCJaJLOL r A headingd@& CJKHaJdOd r Exercise headd$dN5OJQJ\^J6OA6 r Activity headHObH r Activity listd OJQJ^JPOrP r Exercise textd` OJQJ^J\Oq\ rExercise number list & FQ^`QdOd rActivity number list! & F hQ^`Q6Oq6 r Activity text<O1< r B heading6CJ]aJZO1Z r Box A head$%d 'd O Q mH sH POP r Box B head 56CJ\]aJxOx rBox text2d%d 'd O Q `CJOJQJ^JaJmH sH ROR rBox bullet list & FQ^`QBOB r Box display `CJaJ`O` rBox table head!&dP`6CJ]aJZOZ rBox table column head"56CJ\]aJ@O!2@ rBox figure caption#XO1BX rBox figure source/notes$5CJ\aJROR rBox number list% & FQ^`QnO1n r Box title9&$d %d 'd N O Q mH sH ZOaZ r Box sub-title'$dN6CJ]aJjOj rBox table source/notes(&dP`CJaJJOJ rBox table text)&dP0Oab0 rBox type*;<O< r C heading+6CJ]aJTOT rCaption source,d5CJ\aJ`O` rCaption source/notes-d5CJ\aJ2O2 r Case A head.2O2 r Case B head/\O\ rCase bullet list!0 & F hQ^`Q4O4 r Case extract1BO1"B rCase figure caption2LOAL rCase figure source/notes32O2 r Case number4TOQRT rCase number list5 & FQ^`Q8Oq8 rCase sub-title6HO!H rCase table column head7:Or: rCase table head8JOJ rCase table source/notes9:O: rCase table text:.O. r Case text;0Oa0 r Case title<NON r Chapter head=$@& a$ CJ$KHaJ$POP r Chapter intro>d` OJQJ^J^O^ rChapter intro list ? & F hQ^`QbOb rChapter number@&d P ;@(CJaJBOB r D headingA56CJ\]aJ@O"@ rDisplayBd]^^O2^ repigraph"Cxx]^`6OJQJ]^JxOBxr Example text!Dd-D M `#OJQJ^J_HmHnHsH tH u`OAR`rExample bullet list E & F hh^h`jOBj r Example head!Fdx-D M 5;OJQJ\^JZOArZrExample number listG & FQ^`Q`OA`rExample program code H`CJOJQJ^JaJPOABPrExample subhead I`5CJ\aJbOqb rExercise bullet list J & F hQ^`Q4 @4 rFooter K 9r BOB rGlossaryLd`CJaJ4@4 rHeader M 9r O r Key terms boxLNd$d %d &d 'd N O P Q CJOJQJ^JaJXOX r List bulletO & FQd^`QCJaJO r List numberP & F hQd>T.Tf^`QCJaJ.)@. r Page Number>O> r Part headRCJ,aJ,mH sH DO2D r Part introSCJOJQJ^JaJLO1BL r Part list T & F hQ^`QlO"l r Part number!U$xx&d P a$5CJ OJQJ\^JaJ `Ob` r Program codeV]^`CJOJQJ^JaJ8OA8 rQuestions headW^Oq^ rQuestions A headXx$dN 56\]XOX rQuestions bullet listYd OJQJ^JXOX rQuestions number listZd OJQJ^JjOj rQuestions sub-num list#[ & F h^`8Oq8 rQuestions text\0O0 r References]POP rReferences/Further reading^ROR r Table text_dhxxOJQJ^JmH sH DOD rTable column head`5\JOJ r Table headinga56\]mH sH NO"N rTable source/notesbdCJaJ6@26  Footnote Textc@&@A@ Footnote ReferenceH* ?vԂc7:WYfxoyx E * D      !"#$%&'()*+,-./0123456789:;<=>?@ABCDyyyyyyyyyyyyyyyyyyyyyyyyy E * * * * *  } !"#$%&'()*+,-./_0123456789:;<=>?@ABCoxD z z z z z z z z z z z z z z z z z z z z z z z z z z z z z z z z !z "z #z $z %z &z 'z (z )z *z +z ,z -z .z /z 0z 1z 2z 3z 4z 5z 6z 7z 8z 9z :z ;z <z =z >z ?z @z Az Bz Cz Dz&*v - #(/8?)GNW*ahpvN|ւd:0 _P(~x@ {E'n-1^7Q9<DJRY_4f?lrx ^TGf^o w5oT w  IL qHCkh B!K"#$%&'()*+,-W./012 34:567x891:%;U<A=>?@3ABWC&1d&&*p v  @Q0hbOh?"W"%#(w(y,2.d./15678889:?1CyDD;F)GGLLLNQJTT(VVWXiXXZZ5ZGZ]__defBT9TodxISe+ijK\v7Z43TW. X@XpH 2@Rj|!@[n Ar    + S T%\{X78=@JMPS]fopv}  '.56<ENW^elsz{",3:AHOPV`jt{ %,-3459@GNUVWXj "+4=FOX_`afox#*+,1:BJS\elmns| Q p"#$%E'U')++,,--n-0.9.P.b.//00S1}111H22#3w3[6^7777777777 888$8-858>8F8G8S8e8888888888899999)9*9<9=9Q99:: ;;;&;8;C;D;S;^;i;j;y;;;;;;;;;; <<<3<E<W<X<l<~<<<<<<<<<<<<<X=?@@ABCDLDDDEFG5IGIJKLSMOP QSUYGZ]^_```xaabb4fPfgh.h9hshhhijln$o pzpp,qyqqqqrstvix{xxxxx}~ EW`Ɓ؁|]քIޅĆ 4F7Č^1c9T—Pbhz3,|GzϞEơءfҢȣ~8^ èި>jFXm6'Muճ\|w̷)*,-/02356YZlpU0R0@0=0>0>0+0>0' ?0' ?0' ?0' ?000 O0 O0 O000x0x0x0xB00xB0x0xB0xB0x0x0x00x0x0x0x0x0xB0x0x0x0 0x0x0(0x0x0x00x0xB0x0x0xB080x0x0x0x080x0x O0x O0x0@0x0x0x0H&0x0x0x0x%0%0x%0x%0x 0x0x 0x00x$00x0x P0x P0x P0x P0x P0x P0x P0h P0x P0x0x0x0x0x0000x0x0x0xB0x0xB0p0xB0x0xB0x0xB0x0xB0x0x0xB0x0B00xB0x0x0xB0000B0x00B000x0xB0xB0xB0xB0x0 O0x O0x O0&0x0x0x0x 0x00#0x0x0x00x00x+0x0x0x0x+00xB0x0x0x0xB00xB0x0xB0x0B0x0x0x0x0xB0x0xB0x0P0P0xP0x00xP0xP0xP0x0x00x0x0x0x0x0x0xB00x0x0x0xB0x0xB0x0B0x0x0B0xB0xB0xB0x0x0xB0B0x0x0x0B0x0x0x0xB0x0x0x0x0x0xB00xB0x0xB0x0x0xB0x0xB0x0x0x00x0x P0x P0x P0x P0x0000x0x0x0x00x0x&0x0x 0x0x %0x0 0x %0000 0 0 %00 00 %00x0x 0x 0x 0x0x 0x 0x 0 0x 0x 0x00B0x00B0x0000x0x0xB0xB0xB0xB00x0xa0x`0 0 `0 `0 `0 `0 `0 `0 `0 `0 `0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 _0 _0 _0 _0 _0 _0 _0 _0 _0 0 0 0 0 0 0 0 0 0 0 0 0&0x 0x 0x 0x 0x 0x 0x 000x0 0x P0xP0xP0x00P0xP0x0( P0xB0xP0x P0x P0xB00P0P0P00P00x00 O0x O0 x0x0xa0x_0 _0 0 _0 _0 _0 0 _0 _0 _0 0 _0 _0 _0 _0 _0 0 _0 _0_0 0 _0 _0 _0 0 _0 _0 0 _0 _0 _0 0 _0 _0 _0 _0 _0 0 _0b0xb0a0_0 _0 0 _0 _0 _0 0 _0 _0 _0 0 _0 _0 _0 _0 _0 0 _0 _0_0 0 _0 _0 0 _0 _0 _0 0 _0 _0 0 _0 _0 _0 0 _0 _0 _0 _0 _0 0 b0b0xb0xb0xb0x P0x O00 O00P0 O0x O00 O0xP0@P0xP0xP0xB0xP0xP0 O0H O0xP0xP0xP0xP0xP0x0x0x0X0x0x0xB0`0xB0x0x0x0000`0xB0x0xB0x0xB000 O00 O000xB0x0x0x O0xB0x O0B0xB0 O00 O0x00x0x0xB0x0xB0x0B000x0x      !"#$%&'()*+,-B0x0xB0x0xB0x0x0xB0x0xB0x00P0P0P00xP0xP0P0x0x0x0x0xB000 O0 O00 O0 O00 O00x0xB0x0xB0x0x0x O0 O0 O0xB00x0xB0x0xB0x00B0B0B0xB0x00x0xB0x0xB0x0xB0000B000B000B0x0xB0x0xB0x0B000000000B00xB0x0xB0xB0x0 O00 O0x O000xB000B0B0xB0xB0x0x0xP0xB0P0P00000x0xB0x0x0xB0x0B0000000c0xc0B0c0c0xc0B0c0c0c0c0c0c0c0c0c0c0c0c0c0c0c0xc0c0\>0t 0\>0 b0\>0 b0\>0 Hb0\>0 bM0\>0  bK0\>0  00CCCFX %1wzEg8 @ *3HbIIJ=KMTc` _L"$zbdH~JlD  p* "4>.?36;?BGKOPRTUiKxNlU<IV\\\\]]]]^^Z^[^^^^^1_2_{_|_____`.`r`s`````>a?aaaaabbbbccyc)RxֆL5PxxŬ  >.?      !"#$%&'()*+,-./01245789:<=>@ACDEFHIJLMNQS,?,.O]_!!!666888ZZZ5ZCZEZ"z0z2z{${&{}}}LZ\#%ւ%353AC`npvʉ؉ډ۫´дҴ ,:<BPR9GITbdo}Sacguw\jl .0@NPXfhjxzIWY\jlr      P.^.`.0007777778888889999%9'9*989:9 ;;;&;4;6; <<<3<A<C<E<S<U<l<z<|<<<<<<<<<<5ICIEIMI[I]I```xaaah*h,h9hGhIhpppyqqqqqqixwxyxxxx|||}#}%}~~ESUƁԁց|4BDƍԍ֍,.9GIESUP^`hvx')(*Ϟݞߞȟ֟؟ơԡ֡ftvŢǢ~FTV̨ڨܨo:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: $+-0;@F! ::8@0(  B S  ?o*NXp'| }3~³|TD”ZNY`.;]1CmMRbc1p'Rb<0.y%=S3Rb?R5RbAREtSRbcv'U fXRbv[RbGm3(c 4cRb@IfRbKiOFjRb" HnRbksRbnsRb0uuH[nxMT^`.^`.88^8`.^`.^`OJQJ^Jo(^`OJQJ^Jo(88^8`OJQJ^Jo(^`OJQJ^Jo(hh^h`.hh^h`OJQJ^Jo(hh^h`OJQJ^Jo(@^`.@^`.@^`.@^`.@^`.@^`.@hh^h`OJQJ^Jo(@hh^h`OJQJ^Jo(hhh^h`OJQJ^Jo(h88^8`OJQJ^Jo(oh^`OJQJ^Jo(h  ^ `OJQJ^Jo(h  ^ `OJQJ^Jo(ohxx^x`OJQJ^Jo(hHH^H`OJQJ^Jo(h^`OJQJ^Jo(oh^`OJQJ^Jo(@^`.@^`.@hh^h`OJQJ^Jo(@^`.hh^h`.@^`.@^`.@^`.hh^h`.@^`.@^`.@^`.hh^h`OJQJ^Jo(@^`.@^`.@^`.@^`.^`o(()@hh^h`OJQJ^Jo('F>R1Cm<0@BEKv[Zv DKinx0uu" HnFj&]IfX=S34cA<J@Ifnsc1p'tS YLMks?R5~}|m3(ccv'UaF''         Nhe2 W%0i+0*20~Mmh$xN|3}Jrk# lV_+YwP@.0r-O78=@JMPS]fopv}  '.56<ENW^elsz{",3:AHOPV`jt{ %,-3459@GNUVWXj "+4=FOX_`afox#*+,1:BJS\elmns|777777777 888$8-858>8F8G8S8888888888899999)9*9<9=9: ;;;&;8;C;D;S;^;i;j;y;;;;;;;;; <<<3<E<W<X<l<~<<<<<<<<<<<<p@\\MUMS\Brother HL-5040 seriesNe03:winspoolBrother HL-5040 series\\MUMS\Brother HL-5040 series<D S 4dXXA4PRIV ''''} ldArial  ENGLISHBRSP103A.EXEBRRBATOM_5040opt_5040.inicop_5040.iniHL504ini.datHL504def.dat )  Z47 000Ix'RSP103A.EXEdArialdBRLHL03A.DLLEOSC\\MUMS\Brother HL-5040 series<D S 4dXXA4PRIV ''''} ldArial  ENGLISHBRSP103A.EXEBRRBATOM_5040opt_5040.inicop_5040.iniHL504ini.datHL504def.dat )  Z47 000Ix'RSP103A.EXEdArialdBRLHL03A.DLLEOSC X  "(68:<?ACYZ[wxyz#~#YYqqqqqqqqEE E fffff"f#f%f&f.f0f:f;=>?@Anop@pp4@p"pH@p(pX@p.p`@p2p|@pLpPpTpXp@p`p@pzp|p@pp8@pp@@ppL@ppppp@pp@pp@pp@pppp@pp@pp@pp@pp @pp,@pp4@pp<@p pD@p0pd@p4p6pp@pLpNp@p\p@pbp@plp@ppp@pzp@pp@ppp @pp(@pUnknownGz Times New Roman5Symbol3& z ArialO1 CourierCourier New;Wingdings?5 z Courier New"?jD?jD!;Njɩɩ2;?0~MPART IVi2iPerman'                           ! " # $ % & rdDocWord.Document.89q