ࡱ> jM KRbjbj== "BWWTBlppppJJJ8JlKz4LN@Q(QQQRRREGGGGGG$ ΩLkRRRRRk XppQQ4 X X XR$plQQE XRE X X\y4,!Q@N 0r;l|FJST:!$J0z: X! Xpppp CHOICE OF A PRIOR DENSITY In Example 4, the prior distributions of p1 and p2 were given for 17 different values. Note that 0 d" p1, p2 d" 1 and that there are infinitely many possible values between 0 and 1. When practically possible, we give prior and posterior distributions in terms of known densities, such as the Gaussian, binomial, beta, gamma and others. A density is a smoothed bar chart that shows how probability is distributed. An example of a commonly used density for proportions is the beta.  EMBED Equation.3   EMBED Equation.3  Note that the posterior mean of the beta is  EMBED Equation.3 , the posterior standard deviation is  EMBED Equation.3 . Note that  EMBED Equation.3 . Page 201 of your text has examples of several beta densities. There is a different density for each (a,b). The posterior distribution of p1, for a binomial likelihood when beta prior is used: Prior :  EMBED Equation.3   EMBED Equation.3  Likelihood:  EMBED Equation.3   EMBED Equation.3  Bayes Result:  EMBED Equation.3  NOTE THAT THIS IS: beta(a+x,16-x+b) or beta(a+s, b+f) where s = number of successes f = number of failures This is the updating rule for Beta. So we see that a Beta prior updates to a Beta posterior. CHOOSING BETA DENSITIES AS PRIORS. Selecting a beta selecting a and b. Assess the probability ( r ) that a randomly selected cigarette #529 will ignite. This probability will be judged to be the mean of the beta density that is  EMBED Equation.3 . Given the information that the first cigarette ignited, assess the probability ( r+ )of the second randomly selected cigarette #529 igniting. The updating rule says that the updated density is beta(a+1, b) so the assessed value is  EMBED Equation.3 . Solve simultaneously to obtain:  EMBED Equation.3  In Example 4: Suppose that an expert says that he agrees with the given prior mean thus r = 0.057. Also that given this information he would say that his probability of the second cigarette igniting is 0.1. That means that r+ = 0.1 and that this experts prior knowledge has  EMBED Equation.3  Consistency check: Ask the expert to give a value for r- , the probability of ignition of the second cigarette given that the first failed. Using the values of a and b calculate:  EMBED Equation.3 .. Check to see if the calculated and elicited values agree. In Example 4, suppose that an expert thinks that r- is 0.01. We calculate:  EMBED Equation.3  . Can the expert agree that r- is really 0.054? If yes then the beta(1.19, 19.74) density is a good prior. If no, it is not. What to do if the beta(1.19, 19.74) density is not a good fit. We can adjust the values of r and r+ or r- to obtain a consistent result. We can use a more open-minded prior. Also called less-informative. This means smaller values of a and b. The beta prior with a = b = 1 is a flat line. This gives all possible values of the proportion equal probability and is in that sense objective. The updating rule for betas: a becomes a+s b becomes b+f The following is a plot of a random sample from beta( 1.19, 19.74):  The following is a histogram plot of a random sample from beta(2, 20) :  The following is a histogram plot of a random sample from beta(2, 30) :  We will need a beta form of the prior distribution for p2 also. Again, suppose that an expert says that he agrees with the given prior mean thus r = 0.943. Also that given this information he would say that his probability of the second cigarette igniting is 0.96. That means that r+ = 0.96 and that this experts prior knowledge has  EMBED Equation.3  So we can use beta( 2.2, 0.13) for our prior for p2. USING DATA FOR PRIORS Recall our cigarette data for 10 layers. We have data for years 1993 and 2000. If the experiments are very similar, we could decide to use the 1993 data to obtain a prior for the 2000 data and thus combine the information for the two years. We could start with a beta(1,1) prior for both p1 and p2 for the 1993 data. This would result in a beta (1, 17) posterior for p1 and a beta(17,1) posterior for p2. These distributions would now become priors for p1 and p2 of the year 2000. Combining with the data, i.e., 0/24 ignitions for #529 and 22/24 ignitions for # 531, we get: beta(1, 41) for p1 , beta(39, 3) for p2. These posterior distributions combine the data of the two years. The posterior mean of p1 is 1/42 = 0.024, the posterior variance is 0.0235. The posterior mean of p1 is 0.0.928, the posterior variance is 0.0393. What would happen if we simply combined the data, that is say that we have 0/40 ignitions for # 529 and 38/40 ignitions for #531. With a beta(1,1) prior for p1 and p2 we would get beta(1,41) and beta(39,3), i.e., the same result. This shows that using prior in this way simply combines the two sets of data together. An alternative approach to combining data from different but similar sources: Hierarchical Models: Let the probability of ignition of cigarette #529 in year 1993 be p11, in the year 2000 be p12. As in the simple models, we will give pij a beta prior. Let the prior of pij be beta(a,b). The a and b are now unknown random quantities with their own prior distributions. (THIS IS THE MAIN DIFFERENCE) NOTE: We are not saying that the pij are equal for the two years. The beta parameters are given gamma(1,.001) priors. This particular gamma distribution represents vague or objective or noninformative knowledge about these hyperparameters. Gamma distribution: The probability distribution of the parameter a is Gamma((((), i.e.,  EMBED Equation.3  This kind of model, combines the data from the two years in a way which lets the data itself determine how much combining is done. We call this borrowing strength because related data is used to increase the precision of a single experiment. When the data from the two experiments is similar it is combined to a high degree. When it is different, it is not combined very much. One disadvantage of this model is that it does not have a closed form of the posterior distribution. Thus we need to use numerical methods to obtain the posterior mean etc. PROBABILITY INTERVALS In classical analysis we calculated 95% confidence intervals for p1 and p2 to make a judgement of how different they are. In Bayesian analysis we can make such judgements based on probability intervals based on the posterior distribution of p1 and p2. Recall the posterior distribution of p1: Value of p1PriorLikeli- hoodPrior X LikelihoodPoste-rior00.61.00.60.8980.06250.150.35610.05340.080.1250.10.11810.011810.0170.18750.070.03610.0025270.0040.250.050.010.000510.00070.31250.030.00250.0000750.00010.37500.0005000.437500.0001000.50000 Here, 0 d" p1 d" .0625 is a 0.978 highest posterior density interval (HPD) for p1. This means that it is the shortest posterior interval of this probability. Alternatively, suppose that we use prior beta(1,19) for p1 in our example. Then the posterior distribution is beta(1, 35) and P(c1 d" p1 d" c2) =  EMBED Equation.3 . We need to select c1 and c2 so that this probability is equal to some value, say 0.95. A random sample from beta(1,35) gives the following histogram:  Based on the sample, we can estimate that c1 = 0.0 and c2 = 0.08. Note that this interval is quite a bit shorter than the classical 95% CI which was (0.0057, 0.2407). For p2, the posterior based on the beta(2.2, 0.13) prior is beta(18.2, 0.13).  The interval (0.961, 1.0) is a 0.95 probability interval for p2. Again, compare this to the 95% CI (0.759, 0.994) There is an approximate method of obtaining c1 and c2 that is based on Normal tables. This method is useful when both a and b are large. For this method, calculate:  EMBED Equation.3  Then the perc% probability interval for p is  EMBED Equation.3  where :  EMBED Equation.3  Suppose that p has a beta(10,15) posterior. Then r = 0.4, r+ = 0.423 and t = 0.096. Hence a 95% probability interval is: 0.4 1.96 (0.096) 0.4 0.188 COMPARING TWO PROPORTIONS In Bayesian analysis it is also of interest to compare the two proportions p1 and p2 directly, using a probability statement about the difference p1 p2. This requires that we obtain a joint posterior distribution for p1 and p2. A joint distribution gives probabilities for pairs of values of p1 and p2. That is P( p1 = 1, p2 = 2) Is the probability that p1 = 1 and at the same time p2 = 2. Bayes theorem is applied to joint prior and joint likelihoods. For independent samples, the joint likelihood is obtained by multiplication of the individual likelihoods. In example 4, the sample of cigarettes # 529 and the sample of # 531 were independently drawn. If x1 = number of ignitions of #529, x2 = number of ignitions of # 531 then P( data =  EMBED Equation.3 | p1 = p1, p2 = (2) =  EMBED Equation.3  The prior distribution needs to be elicited jointly. It is possible that an expert will consider the prior knowledge of one proportion not relevant when the prior knowledge of the second proportion is being quantified. In that case, the prior distributions would be independent and obtained as a product of the two prior distributions. In Example 4, if the expert apriori considered his knowledge of p1 and p2 to be independent then he could use the product of beta( 1, 19) and beta( 2.2, 0.13) densities.  EMBED Equation.3  for 0 d" p1 d" 1, 0 d" p2 d"1. If the prior knowledge is not considered independent then the prior has to be elicited jointly, that is, we have to obtain probabilities for pairs of values of p1 and p2. Again, the probability distribution could be made discrete for simplicity. An example of a possible joint probability distribution is: p1 00.06250.1250.18750.250.31250.3750.43750.750.81250.8750.10.050.93750.10.10.10.0510.30.2 Applying Bayes Rule to joint distributions is straightforward. If both priors and likelihoods are independent then the posterior distributions are also independent and the two proportions can be done separately. In example 4, we get the posterior distribution of p1 and p2 as the product of beta(1, 35) and beta(18.2, 0.13) In cases when the p1 and p2 are not apriori independent we apply Bayes Rule to the joint distributions. In example 4, we have likelihood: P( data =  EMBED Equation.3 | p1 = p1, p2 = (2) =  EMBED Equation.3  prior: P1 P200.06250.1250.18750.250.31250.3750.43750.750.81250.8750.10.050.93750.10.10.10.0510.30.2  Likelihood: P1 P200.06250.1250.18750.250.31250.3750.43750.750.81250.8750.0420.0140.93750.35610.1270.0420.013110.3561 Prior x Likelihood: P1 P200.06250.1250.18750.250.31250.3750.43750.750.81250.8750.0040.00070.93750.03560.0130.00420.000610.30.071 P(data) = 0.429 Posterior: P1 P200.06250.1250.18750.250.31250.3750.43750.750.81250.8750.0090.010.93750.080.030.0090.00110.70.16 The set  EMBED Equation.3  has 0.97 posterior probability. We can also obtain the posterior probability distribution for the difference p2 p1. p2 p110.93750.8750.81250.75Prob.0.70.240.030.020.01 So the interval 0.875 d" p2  p1d" 1 has posterior probability of 0.97. 2.3 MODELS FOR MEANS Recall the SRM 1946. For lab #1, the data consisted of mean concentration of PCB 101 calculated from 24 observations. This is a type of data for which the likelihood function is usually represented by the Normal distribution. The justification for this is the following result: Central Limit Theorem: If  EMBED Equation.3  is an average of a large number (n) of independent observations which have the same mean m and standard deviation h, then  EMBED Equation.3  has the Normal distribution with the same mean m and standard deviation equal to  EMBED Equation.3 . In fact, if n is large enough, we can use an estimate for the standard deviation (sample standard deviation s) and still have the normal distribution. We use this fact here: In the SRM 1946 example, lab #1 produced a mean value  EMBED Equation.3 . The sample size for this lab was n = 24. So it would be quite reasonable to assume the Normal distribution with h= 0.7 for the Likelihood function. That is, given that  EMBED Equation.3 , the likelihood of m = m* is  EMBED Equation.3  We can now use this formula to calculate the likelihood for different values of m*. PRIOR DENSITIES FOR MEANS When no prior information about the mean is available, it is common to use a flat line. In this case, the posterior density will be a normal with the mean equal to the sample mean and the standard deviation equal to sample standard deviation over (n . In the SRM example, this means that we would use Normal( 38.1,  EMBED Equation.3 ). When we wish to use an informative prior we generally use a normal density. That is, we assume that the function  EMBED Equation.3  represents the prior probability. The parameters m0 and h0 are the prior mean and standard deviation of m. The parameters m0 and h0 are generally elicited from an expert. A noninformative form of this distribution is to assume that m0 = 0 and h0 is a very large number (100 times the sample standard deviation). The Updating Rule for Normal Models: For n observations from a normal(m,h) distribution with average  EMBED Equation.3 , if the prior density is normal(m0,h0) then the posterior is normal(m1, h1) where  EMBED Equation.3   EMBED Equation.3 . Probability interval for a normal mean: A perc% posterior probability interval is  EMBED Equation.3 . For example, for lab # 1 in the SRM 1946 example with a flat prior: 95% posterior probability interval is 38.1 1.96 (  EMBED Equation.3  ) Comparing two or more means As with proportions, if there are several means to be compared to each other then we need to obtain joint likelihood and joint prior distributions. In most cases independence will be used to justify multiplication of the individual likelihoods and prior distributions. In such a case the following result holds: Rule for Differences: If the posterior densities of mA and mB are normal(mA1, hA1) and normal(mB1,hB) respectively, then the difference mA mB has a normal( mA1 mB1,  EMBED Equation.3  ) density. A perc% posterior probability interval for the difference mA mB is: mA1 mB1 zperc  EMBED Equation.3  Example: Suppose that you wish to compare the means of the measurements from lab1 #1 and #6 of the PCB 101 data set. If we use flat priors for both, we get posterior normal(38.1,  EMBED Equation.3 ) for lab #1 and normal( 39.3,  EMBED Equation.3 ) for lab #6. Then the posterior distribution of the difference between lab #1 and lab# 6 means is: Normal( -1.2, 5.152). The 95% posterior probability interval for the difference is: -1.2 1.96 (5.152) . Hierarchical Models for Means. As in the case of proportions, it is possible to build relationships between different data sets by using hierarchical form of the prior distribution. Example: SRM 1946 PCB 101: Lab IDMean Conc.St. Dev.# obs.138.10.724234.50.33331.50.56430.81.696532.52.596639.323.0420 Hierarchical Model Data: for each lab i,  EMBED Equation.3 , si , i=1,,6 Likelihood: for each lab i, the observations are normally distributed with mean i and standard deviation i . Priors: The means i have normal prior distributions, that is they are normal(m0, h0). The m0 (the consensus mean) is the common mean across the labs. It is unknown and has a prior distribution, that is normal( 0, 10000). This type of model combines the data across labs. The following table gives the results for PCB101: Summary of Results: TypeConsensus mean95% CIBayes34.41(30.95, 37.54)Grand Mean36.50(30.86, 42.14)Mean of Means34.45(30.73, 38.16)MLE34.59(32.05, 37.14) Some Comments about hierarchical models: They combine alike data more than different data. They will combine apples and oranges if you set it up that way. They borrow more, that is give more weight to similar data when sample size is small than when it is large. PAGE 45 PAGE 55 46 ./013`atuvw~qjCJ4EHOJQJU3j@ B*CJOJPJQJUVmHnHphujCJ4OJQJU jEHU3j7@ B*CJOJPJQJUVmHnHphu jEHU3j7@ B*CJOJPJQJUVmHnHphu jUCJ4H*OJQJ CJOJQJ CJ4OJQJ"68m23`z$d-DM a$d-DM ^$d-DM ^a$ d-DM )RFRJRgh  ! Ƙ~w]V j(EHU3j8@ B*CJOJPJQJUVmHnHphu j EHU3j7@ B*CJOJPJQJUVmHnHphuCJ4H*OJQJ jEHU3j7@ B*CJOJPJQJUVmHnHphu jU CJ4OJQJjCJ4OJQJUj!CJ4EHOJQJU3j@ B*CJOJPJQJUVmHnHphu  % e f g " # $ % $d-DM a$ d-DM ! " # $ M N a b c d g t u v k l  l m vogCJ4H*OJQJ jBEHU3j8@ B*CJOJPJQJUVmHnHphu j<EHU3j/8@ B*CJOJPJQJUVmHnHphu CJ4OJQJ j6EHU3j!8@ B*CJOJPJQJUVmHnHphu jU j]/EHU3j8@ B*CJOJPJQJUVmHnHphu @    H d-DM hd-DM ^h & Fd-DM   DE   HI|}ҩҩngddҩCJ4 j{]EHU3jWx @ B*CJOJPJQJUVmHnHphu jTEHU3jw @ B*CJOJPJQJUVmHnHphuCJ4H*OJQJ j MEHU3j8@ B*CJOJPJQJUVmHnHphu CJ4OJQJ jU jGEHU3j8@ B*CJOJPJQJUVmHnHphu&DEGH[\]^@Avw}~vҿҲҪҗ}vҗҗҗҗҗҗҗҗҗ jzEHU3j9@ B*CJOJPJQJUVmHnHphuCJ4H*OJQJjtCJ4OJQJU6CJ4OJQJjn6CJ4OJQJUjhCJ4OJQJUCJ4H*OJQJ CJ4OJQJ jU jbEHU3jy @ B*CJOJPJQJUVmHnHphu.2rs-.CD/d$d%d&d'd-DM NOPQ d-DM DFH#$2_`CD$d-DM a$ d-DM ^_123YZ$d-DM a$ d-DM vwN\$&GI!"5678dfrsyz4 6 𥘿𑉑CJ$CJ$H*OJQJ CJ$OJQJj]CJ4EHOJQJU3j@ B*CJOJPJQJUVmHnHphujCJ4OJQJU jbCJ4OJQJ j,CJ4OJQJ jaCJ4OJQJ>*CJ4OJQJ CJ4OJQJCJ4H*OJQJ4!:;dfgh|} d$If d-DM  STS$$Iflrdc$` 04 la d$If    * 4 6 B J X h S$$Iflrdc$` 04 la d$If h t v SS|$$Iflrdc$` 04 la d$If 6 t v @!B!h!j!!!!!!!L"N"^#`#####$$$$$$$$$0$1$2$3$9$<$$$$$$$$h%i%%%%%T&U&[&\&&&&ﴬjCJ4OJQJUjDCJ4OJQJU6CJ4OJQJj6CJ4OJQJU jEHU3j @ B*CJOJPJQJUVmHnHphu jUCJ4H*OJQJ CJ4OJQJCJ$ CJ$OJQJ; !! !2!S$$Iflrdc$` 04 la d$If 2!@!B!N!R!`!d!h!j!x!SPST$$Iflrdc$` 04 la d$If x!|!!!!!!!!!S4$$Iflrdc$` 04 la d$If !!!!"#$w$$SFFFFFF d-DM $$Iflrdc$` 04 la d$If$$$b%%%%&&'&&&&&&''1'2'R'''''''''' ( d-DM &&&&''-'.'/'0':';'N'O'P'Q'W(X(^(a(((((((((((7)8)>)?)M)** ****Ƭƅxpppphppppppp>*CJ4OJQJCJ4H*OJQJjCJ4EHOJQJU3j9@ B*CJOJPJQJUVmHnHphujCJ4EHOJQJU3j@ B*CJOJPJQJUVmHnHphu CJ4OJQJjCJ4OJQJUj{CJ4EHOJQJU3jb@ B*CJOJPJQJUVmHnHphu( (((((I) **+ +"+++,, -V-X-b-d--.9.:...W/X/0 d-DM **R*T*\*^*****f+h+++,,--x-z-------------------...//// 0 0»vjCJ4OJQJU jEHU3j>@ B*CJOJPJQJUVmHnHphu jpCJ4OJQJCJ4H*OJQJ CJ4OJQJ jEHU3j@ B*CJOJPJQJUVmHnHphu jUCJ4OJQJehr CJ4OJQJCJ4H*OJQJ.0 0:0p0110212m22222222222 d$If d-DM  020406080L0N0d0f0111122222222233#3$34353A4B4H4I4444455(5)5*5+5.5/5666 6 66CJ4H*OJQJ CJ4OJQJ j7EHU3j@ B*CJOJPJQJUVmHnHphu jUCJ4OJQJCJ4H*OJQJjCJ4OJQJUjCJ4EHOJQJU3j@ B*CJOJPJQJUVmHnHphu CJ4OJQJ1228$$Ifl T$ d40$$$$4 lal2222222222 d$If 22@$$Ifl T$ d40$$$$4 lal2222222222 d$If 22X$$Ifl T$ d40$$$$4 lal2222333333 d$If 33t$$Ifl T$ d40$$$$4 lal333333 3!3"3#3 d$If #3$3D$$Ifl T$ d40$$$$4 lal$3&3*3.3/30313233343 d$If 4353$$Ifl T$ d40$$$$4 lal5363v3w3 4 4~4444 5 6P666666666667 d$If d-DM 666 6"6H6J6L6N6P6\66666667777<7>7h7j777777777777777+8,898:8I8J8b8c888888888888889999D9E9W9缶缶缶jCJ4OJQJUCJ4 H*OJQJOJQJ j.EHU3j>@ B*CJOJPJQJUVmHnHphu jU CJ4OJQJCJ4H*OJQJ jpCJ4OJQJB778$$Ifl T$ d40$$$$4 lal7 777777777 d$If 77@$$Ifl T$ d40$$$$4 lal7,7.70727476787:7<7 d$If <7>7X$$Ifl T$ d40$$$$4 lal>7J7L7T7^7`7b7d7f7h7 d$If h7j7t$$Ifl T$ d40$$$$4 lalj7x777777777 d$If 77D$$Ifl T$ d40$$$$4 lal7777777777 d$If 77$$Ifl T$ d40$$$$4 lal777777778 8888$8+8 d$If d-DM +8,88$$Ifl T$ d40$$$$4 lal,8182838485868788898 d$If 98:8@$$Ifl T$ d40$$$$4 lal:8A8B8C8D8E8F8G8H8I8 d$If I8J8d$$Ifl T$ d40$$$$4 lalJ8P8Q8W8]8^8_8`8a8b8 d$If b8c8$$Ifl T$ d40$$$$4 lalc8j8q8w8}888888 d$If 88H$$Ifl T$ d40$$$$4 lal8888888888 d$If 88$$Ifl T$ d40$$$$4 lal88888888888888 d$If d-DM 888$$Ifl T$ d40$$$$4 lal8888888888 d$If 88@$$Ifl T$ d40$$$$4 lal8888889999 d$If 99h$$Ifl T$ d40$$$$4 lal9 9 99999999 d$If 99$$Ifl T$ d40$$$$4 lal9%9,92999@9A9B9C9D9 d$If D9E9L$$Ifl T$ d40$$$$4 lalE9G9K9Q9R9S9T9U9V9W9 d$If W9X9$$Ifl T$ d40$$$$4 lalW9X9Z9j9k9v9x9y9{9|999999999:::::":#:6:7:8:9:Y:\:::::::::::::::<<<<^<`<y>z>>ҶҝҝҝҝCJ H*OJQJ CJ OJQJCJ4H*OJQJ^Jjv[CJ4EHOJQJU^J7j@ B*CJ4OJPJQJUV^JmHnHphuCJ4OJQJ^JjCJ4OJQJU^J H*OJQJOJQJ CJ4OJQJCJ$CJ45X9Y9j9k9v9w9z9}999999999 d$If d-DM 998$$Ifl T$ d40$$$$4 lal9999999999 d$If 99@$$Ifl T$ d40$$$$4 lal9999999999 d$If 99`$$Ifl T$ d40$$$$4 lal9999999999 d$If 99$$Ifl T$ d40$$$$4 lal99999::::: d$If ::H$$Ifl T$ d40$$$$4 lal:: :::::::: d$If ::$$Ifl T$ d40$$$$4 lal:::":;:[::::::::: d$If d-DM ::::::::I|@@@@@@ d$If$$Iflֈ: 3SSSSSS04 la:::;<b<I<<<< d-DM $$Iflֈ: 3SSSSSS04 lab<<)>*>^>_>v>?<@=@ A!ANAkAlAAAAAAB d-DM 3hd$d%d&d'd-DM NOPQ^hhd-DM ^h>>>> ? ?? ?!?"?u?v?????s@t@@@@@5A6AǾǾǾǾwhǾǾL=ǾjoCJ4EHOJQJU^J7jɝ@ B*CJOJPJQJUV^JmHnHphujiCJ4EHOJQJU^J7j@ B*CJOJPJQJUV^JmHnHphujeCJ4EHOJQJU^J7j@ B*CJOJPJQJUV^JmHnHphuCJ4OJQJ^JjCJ4OJQJU^JjaCJ4EHOJQJU^J7jܜ@ B*CJOJPJQJUV^JmHnHphu6AIAJAKALAlAmAAAAABB3C4CGCHCICJCCCCC̿l]A7ju@ B*CJOJPJQJUV^JmHnHphujCJ4EHOJQJU^J7j @ B*CJOJPJQJUV^JmHnHphu jCJ4OJQJ^JjIxCJ4EHOJQJU^J7j@ B*CJOJPJQJUV^JmHnHphujCJ4OJQJU^JjsCJ4EHOJQJU^J7jɝ@ B*CJOJPJQJUV^JmHnHphuCJ4OJQJ^JBBNCOCCCCCCCCJDKDDDEAEBEEEFF FIFJFuFFhd-DM ^h d-DM CCCDDDD[D\DbDcDDDDDEEEEEEEEEEEEEEEEFFFFFFFڴډzjCJ4EHOJQJU^J7j@ B*CJOJPJQJUV^JmHnHphujCJ4EHOJQJU^J7j@ B*CJOJPJQJUV^JmHnHphuCJ4H*OJQJ^JCJ4OJQJ^JjCJ4OJQJU^JjMCJ4EHOJQJU^J%FFFFuFvFFFFFG GGGGGHHHHHHHHHHHHIII IIII!I#IǾǾǾǾwhǾ^^^^^^^^^^CJ4H*OJQJ^JjCJ4EHOJQJU^J7j @ B*CJOJPJQJUV^JmHnHphujJCJ4EHOJQJU^J7jHU@ B*CJOJPJQJUV^JmHnHphuCJ4OJQJ^JjCJ4OJQJU^Jj7CJ4EHOJQJU^J7j;@ B*CJOJPJQJUV^JmHnHphu$FFFFF"G?G@GGGyHzHHFIIIIIJJJK6KKKKAL$d-DM a$ d-DM #I$I7I8I9I:IIIIIIIIIIIIIIIIIIIkJlJJJJJJJJ;鴪͛pj3CJ4EHOJQJU^J7jHZ@ B*CJOJPJQJUV^JmHnHphujYCJ4EHOJQJU^JCJ4H*OJQJ^JCJ4H*OJQJ^JjCJ4EHOJQJU^J7jDW@ B*CJOJPJQJUV^JmHnHphuCJ4OJQJ^JjCJ4OJQJU^J JJJJULLL M MMM M!M$M%M1MLMNN8N:NR?RARBRCRJRKR0JmHnHu0J j0JUCJ4OJQJ^JCJ4CJ,(R)R5R6R7RCRDRERFRGRHRIRJRKRh]h&`#$ d-DM 1h/ =!"#$%Dd$B  S A? 2/X39RDo`!/X39R$ u2HH CxڕkQ}3Q\׀)\AdVNJE[ "D Yi`%36vZ؉"51Mf]v ̼{߲g(T|`#U=hۅs棢IvP94o}\T6!sky-O&5ͮ[ *NNOanz?ϫoQ3YqshlC^#Y2~d_kF [T_Q.ZKIRu[eGe}nZv^^Wڼ4.eBeSu6O5T1S{Kuu:ݩu~R=j%~ZPY}Ǟ(UUW*֙3TV,efV%C}&7eܻv%*=SyFVfq\H=\C#P<>Q $ BHR u2HH xVkVɎؖc/ ).=}C 4m R E+9Ҳ7mɡcK T ͵^?Rrsg>ޖ }3{f}CF,_,Hf|]Xh0u:oG}ܫPyi Q40w )T> 帺S0S|QjA6znP44e MesA9d &=C.B lf_۷2Y췄[nShMN`p?Ԝ;.d):@1%G KYBҊOsK rJ*@:,hZ5i .:Jr]\Zy"_lFȺ!=(|BY1?9z1;t*: kpqa:[l,Yt(T:?x}3zaYV6SIuGI.2L?`n1͙l?:ic!ېv'ht(]ܧ^rv wNsN48ЊQ^_IemA(Ln~')^,r1Aiv9Y㓏2{!l Z(V mvfAS DqAPQP4|oIn!#2@NŎYap 7 y:,J޼M&s,JMkF֕8 m;Mq[s(.JǠL#ڶZUm1X fӵufaVS7Ȯx?BQN _CK{djUNn ?An!4Kx U3/3EʇvCU9'WkI WUzJO-]V#^scr8*$UVr&C'H g Z:jR&WZYGrNGq}Ri#8# -#OV4xV|tah?wP+31OdmTIJf \2ٜj2qJ2vY'7c|Cw}Gv;/e`Nk~4zㆸS~ܔNg[xZOR;S⬅ys<;KW`rJ.?9%: `05*=C$^3K Dd $B  S A? 28 oAc`![\{W 7dJhW7fV50[E bxcdd`` @c02tY,P1=F (`d?3%$ lʛ27 27)?ȉeZ> X.Y1Lc0B2sSM•4F0DPS_r>.7BLe4 u-T-rb]^4jBO|(1b# u.`lbskn5IlT`@VPd[@ iEĹ!DՄX03ĆR`)i5jJ[?10X= 3yl C8 oA50[E x}TMoTLCmeZD%zXU̠F XdSE9qGghZ)RE4 .+X?v}3AO̻s{onpA1r\d=Z} "+~͟dOq%&߈SVW4hm` C!sO*l]ُF<'Ugo)^&lyCR\u 6}ٵ7akOQ 5C1-أRu%+J2b Z[YUxNIM)պZ"7oʂ!x>V~{猺tGy-{S'!_!?Kǂfic>9Ą4rLNCwOc˖E%y]m{k}+p1<9*( IzAyEɑ$ǹ\Oǧ[xRPeMX{.ðyE4;-M$^҂Os>{$^U9F!ӱ&&ZH ,*GCH˙R)W!e@$y<s^ڙ,,%»Ê8gtʋi/CuwujC{@2+ޗGx:[s&s&:Za7q@kjk`cK6t]wXВ\ h Zt#6 C E,{Хe i7;PD]ǡ˖ˍёY7ݣMli-nvH9`ۂ'eJ"C!clXV0y,̨€!DC 'VEszyuStUPZeˮR| k7ER=s 6_ %( 䚢 05]u7.p/ {Dd\B  S A? 2glBRT`!d`o|[X3[|֢w+-HU xcdd``~ @c02hY,P1=F (`d?3%$ lʛad3|-wr.rs{~  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~  Root Entry F;l@HData WordDocument"BObjectPool10;l;l_1074673560F0;l0;lOle CompObjfObjInfo "%&'*-.14569<?BCFKNQTUVY\]^_behklorstux{|FMicrosoft Equation 3.0DNQE Equation.3n.39q beta(a,b)=Ga+b()Ga()Gb()p a-1 1-p() b-1 ,0p1Equation Native _1074673573J F0;l0;lOle CompObj fFMicrosoft Equation 3.0DNQE Equation.3n.39qd Ga()=t a-1 e -t dt 0 FMicrosoft Equation 3.0DNQE ObjInfo Equation Native  _1075103981YFЎ;lЎ;lOle CompObjRObjInfoEquation Native :_1075104253FЎ;lЎ;lEquation.3 aa+bFMicrosoft Equation 3.0DNQE Equation.3`  aba+Ole CompObjRObjInfoEquation Native |b() 2 a+b+1() FMicrosoft Equation 3.0DNQE Equation.3n.39q~ beta(a,b)p a-1 1-p() b-1 ,0p1_1074673588FЎ;lЎ;lOle CompObjfObjInfoEquation Native _1074673654 'FЎ;lЎ;lOle  CompObj !fFMicrosoft Equation 3.0DNQE Equation.3n.39q Ga+b()Ga()Gb()p 1  a-1 1-p 1 () b-1 ,0p 1 1ObjInfo!#Equation Native $_1074673665$FЎ;lp;lOle (FMicrosoft Equation 3.0DNQE Equation.3n.39q p 1  a-1 1-p 1 () b-1 ,0p 1 1CompObj#%)fObjInfo&+Equation Native ,_1074673689",)Fp;lp;lOle /CompObj(*0fObjInfo+2Equation Native 3FMicrosoft Equation 3.0DNQE Equation.3n.39q 16!x!16-x()!()p 1  x 1-p 1 () 16-x_1074673697.Fp;lp;lOle 7CompObj-/8fObjInfo0:FMicrosoft Equation 3.0DNQE Equation.3n.39qa p 1  x 1-p 1 () 16-xFMicrosoft Equation 3.0DNQE Equation.3n.39qEquation Native ;}_1074673711h3Fp;lp;lOle =CompObj24>fObjInfo5@Equation Native A_10746738518Fp;lp;lOle Dy p 1  a+x-1 1-p 1 () 16+b-x-1FMicrosoft Equation 3.0DNQE Equation.3n.39qCompObj79EfObjInfo:GEquation Native H@_10746738736@=F;l;l$ r=aa+bFMicrosoft Equation 3.0DNQE Equation.3n.39q= r + =a+1a+b+1Ole ICompObj<>JfObjInfo?LEquation Native MY_1074673903BF;l;lOle OCompObjACPfObjInfoDRFMicrosoft Equation 3.0DNQE Equation.3n.39q a=r(1-r + )r + -r,b=1-r()1-r + ()r + -rEquation Native S_1074493203GF;l;lOle WCompObjFHXfFMicrosoft Equation 3.0DNQE Equation.3n.39q a=0.057(1-0.1)0.1-0.057=1.19,b=1-0.057()1-0.1()0.1-0.057=19.74ObjInfoIZEquation Native [_1074493527E^LF;l;lOle `FMicrosoft Equation 3.0DNQE Equation.3n.39q7 r - =aa+b+1FMicrosoft Equation 3.0DNQE CompObjKMafObjInfoNcEquation Native dS_1074493864QF";l";lOle fCompObjPRgfObjInfoSiEquation Native jEquation.3n.39qg r - =1.191.19+19.74+1=0.054FMicrosoft Equation 3.0DNQE Equation.3n.39q_1074674122;cVF";l";lOle mCompObjUWnfObjInfoXp a=0.943(1-0.96)0.96-0.943=2.22,b=1-0.943()1-0.96()0.96-0.943=0.13FMicrosoft Equation 3.0DNQE Equation Native q_1075105232[F/;l/;lOle vCompObjZ\wR   _` !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^kacbdefhgilmnopqrstuvwxyz{|}~-Č`i iEVFD~A$6f\6 'of3Hvh.bpS3py n 0p3y'>CglBRTJ+-HU rx}Sn1M_miU6$l !@ޑMWث&*qw@}(eloYy=3YyF 5&4튓Y^Icw jj{V7) hR\\B0N taVj>G@<|f477<h>Gs|<PCckBy6Pn}{jy[SŦ/ N e&|Ht!N1 Ϣ3W]J]029= ;7gi4¶Ix:K?Ocr~aA0<Ʊr-1)L(JrD[DL7C `*b>[, 6C 2cE2hW-MŔe˩$NoЯӂGZsG2\dn%vhpG+rgϷ0 2'tUR/dy*q6 n8Lnjj1K 1JJPnي \Ks lS,K#L5w^7$5;3 |WpaLOmEe\r'NRêW_.վ4.)\/-DdX B  S A? 28 ,H xk=e`!  :*ЭNm 4p yxڕJAϙlY/QI!+xA"Fm A0/A"x-,D0Z0?+fef0Ly̙@dZGE^Zsh=l|jDrXmA3U>ԫ.Y@e^;NcAq.;ktXB;UTiFLUD#? "aۥ@v\iR߈CUU}.Y՗X3P-,[k4)3AkG)9ڔORVsvRTXeX[RVs6슺EyUxCnln a΃~ezkx_13~<q\g'C ,H xk= 4p -xVKoFRNR^?8 %>z />h) SJү9(rh9 >CH;ˇ(A ܙofv)n].a[^k.区~1l~6Kqu]`#Gĸ+`qq>K8g 1"ܿuN pVt-Gwm4ah.F_OGI"byM V7&bTGk)*v&.|YV7PSKJ^J*yZL3+kttk'{7_ƽLX㟩dc\|z90^h$xvѫmC$ ~3yi8FWQ l"!bɉ'\~R F3ɾMy3)'PeXDM髄f[ f,g6P(Y:EieeR*xU$b惬"y_f~`g/vzj0JˋSLKZKeTSHL %ūh$Nq>s>)CMhAcmMn-B?7K ?YDd4B  S A? 2+/!2[ 3o`!/!2[ d)9(<xڕKPI#+"b7'Mwb]tgɿIpPCn jsVc"t*p'!5ULkýQm +K5j/6_Y7ez 3EV.'DtcEI[/qE$< WlV\Aql˲H50ĺ:$̺:E59og[\YP uu5R͵@Ԩ uuMzErme]ӠXc]Ә 659d*>smg]r7qM Ù~K rWO#Tɺsԍ{R 7bMl%]7ڠ䔻3•_||r@}Oi\CݦsO9zxR)9(<x}UMd{ӆݍnب| kqP^`ލؖJCVkGPCN\8qڱZd}fy4~| p(_(?~<Z7l=| {eX}, Hfq4 3%N~pNނ'9Ka!(t sKtΟ'RDDPS8ړz[$`nb>f@oC.WP*r$\SK.CL^93tH;X%(jq_k!!vKK!.񾦆;?;Uۧ]Z.u-]o"tܧw% 5]W(,%G7?I%Cq- 'S?1;[1f3Lc )Y*25b:V3aN}oƶ3f3/cH![ae{" O34pNz Y2 OvM?/ bxQe,w t/ O (ȓ< G!FLMP^e_C-*{[GJ@/E _)4# d| =5.&;tEGmLjI-\kon2U:C%arPyshUC!hf9N{m(:ŌL!AdLZoDh$Ъё f/[Rz"j/'_ Ifu^{򳮸(?S嫆|w?/?ގD`Z /T(HsM|IL =4}}+Dd\B  S A? 2|N*Oq o`!i|N*O +2 9 xڕOQǿQ 46abj:0 &&.M !*NlAcdy)_.r}߷jtltrg'Q4JZ F5&z' j027xprjDݶp>}ݨ/C6Go2R}6]jQQʸ~cSĪETւZt*ַ*kq7*A ݒ$ר:$A;*k?.Y;eQEg훖_0VT:8Pӻ7CI,VY=&GܕT|O%ߥ_5_@uumķW93ڑΚLj9oqawMC t[pk!N +2 9 xVoTviR)QڍHڽ@%.+!DZۨe"A{@B\ qBJHg>aɉg7FҤ6ِk.^Hj|Kd퇳 &Y9?A/Dq'/  BY!7>8!!M wHSlxkui7Ici0"6HUE_=$$UzPB"++R$` U_dK\GjUl=8O2S17kme +,Vw7n#Se=&UWqav?=n!t{B[;S V{&$mIM(uiI9g5'5 Q r;w{sQ|?pcƧ2: Chp(aZ{B%w :Co28A4tb:5ya+pQEA/hxL#/tWI L 15itxqH5n@6d0xA׉酂C/84wNAp @x syV<4>ay*崙spri@0STID|$VZىЁOXŇEEIi^G~)T "/:p#D%CJ ~ǧ^I r^%;Êkv$+kQYT7&ar.Nwcg8pӢi$=H,^B\SYp4jFf6JŘi ^1]բjZKKm,ECUett<\*OH'a`mUpy٢]nPf"36 lAW4f/BS ?8Q-Γ@)7]x6I͹2`'A3s5 \sha&uY]hF}\׊a|#!,#'*jUKa¸2 M&.h[\L d6mr7@- }/M5O\ޛ¯]DUI~R۞^EKG~޼qS~ܚeU帲.[|{Nȭq}$x 6.X6.VYyIUekY^gɛzfOY7{>Ɓgj:xsט7Z^}Q>Cg )/CdӄP.`mx}UMT|uqM( Xdg;Bf $83V۲4**;4겨 TlT ~}ϱ '9ys9ן[?X 庒b C,IiEG| T=#ߩ-h$My'q NWA<X>:'`!-8 <#>?}%QC lo(ʷXd5Jձ0>DtmS(*e!*cBvWKr{op4ȵ=DiAo4}KT/mm)ץkSaٔUͳMފggG ֑ĽvỚr?Cm0s0ZhaaFMSBslj3 g7ԋstleu|/ǹ6$8i~&ѐ10:$qtW589_5J%y:FS1 qh(GťBi'ʥ/b8F h yA:~PN҅` .#STS\΃Dzde8kqvqV 򿞅fcWW\ɂ uuv\жt:]X@kUS[Q&g6 XL,Nw{h:-_U/soR!ѱgUw -9a}S#)]J ׫m!uH*11zܢ׭zPl}]EO;y1Biԥcs[@{kq,,m%GC\Rу5_ˇ|A_ (u޳d9]Y p{W{m47ذ^KAU~wPū"I|Dd8B  S A? 2 4X WN/`!q?0!Ͽ6)v xڕ;KALDHD "&6"`#`!$($6j|wvv 6v#ha -;ͺ$.=o5h f$0# ?mzh\j/dmQ;ZXkijۿxMgMmDaDWwm#o};{  .^9⼫Pg[G -Ns(ɵR.yeȳ\rq/A.k\-Aj|>ڷ!و*# R4˭Ȑ3,ZĽԃ79gigGZYkiü t%eXc?uϪaUsb=Fu{*rYUX"c.ֹk׏qY#mؾ0.'vvp^dXPf'絧DB+(aSF"N#I&ʮbWK+SGPm_tk{Ɠ,y~;=g3[@sX//sS\_@R\ޟ'p $i7ŚGi \!GqMXَAS [0 Vݛo*hG]6e Є6(ҐИ}Mm|E߁YXi/{K-W,5l]|EgE[RőBg3F“ &,$ΰ&ֿ!ɠ in `Qd ?zYvY2IT9;#5:<({Dm|,2D?0R&+"+(9DФ X[4Pf/-brJy{1svLtYF~(X1t&ˎX#2 =fc5r0^/S/z^0nQQ?dрeAދ$"Tr ឩ]OZV;Cl Z(!.DTVtu1F%԰'IhRYXH# KB3g8rXArR?mjԹtvr䖀ZLx2Muķ8]-ٶe;˸!k l`9vG1Wp{׋|2D3v- cô10@!( ]>tt9x: (B.bUvhI[Q:>ze425^:؆4w*T6+1T1vL 2l!ZcZ\;%d9%f q35|`jܩ*h98rcF_9y FsڎQi&Mש]V> /}_/׿A9iJ\,Tr iV {yS~noT>L-I0OLX%R * ʟ\_'+< #/Dd 0B   S A ? 2w:Aa$7`!hc/#v.D.& E`mBxcdd``db``baV fxd,F^equ02Ȓcjl@?H @fzM\PU|30I] H+22lee`0 M-f 6m3391HMٍXˍS2HCTn=O[+xT'6VQ#b ĺUhQtknŖ?QtEb  Jbݪ  Oȋ'嗈s9'C6SXˈ"bSeS/'CZ R?3ĵCw:Aa2`mJx}TKG+^#ٗ@`,$!$_=af$kQ ^ s#LrqS4UW_tM X5 hƕTGַhh РrROZwe()IjHpn~w8Kc<}x$bDho }7N(ۇ`k) qiчCBȊ2{>D=~T޻pIp(LP\2sCuU}KO7ynlѣ_ T?s3yx&Σdo!KQ=^\Ӏj\IO9~12hCs= 9v)OyHI\hq!8.ܽlL.hp albvطdsö 8VR1/- :Y?T_d̎҅7gJzVXz 9nRzU.A>bl`CYJ}Uopԧd<,L8N':a+[3D&&ZDEįEhQyTFJQ" 8Y@VxEJy! WQ"r9hMs\-j,x`~<)]~c4ks e=O{ZUٲns3&|&Sn?5 54ǢWIN2Eu6d6k mgP|ϯ1W \]t+,7Pzt"s=$h[CBRkz@ٔ^v[RFTswYSZvV" zp}eA6:%aQ .\՞Jf[1_9o&}ɧ :=-SGn+ urtݣzr(DJW_ IvD|p-HDdB  S A? 2)9y -%ך2*7Co`!9y -%ך2*tH- xcdd``bd``baV fxt<,F^equ02Ȓcjl@?H @fzM`d0B2sS;t2s{~-Č`i iEVF"F @ܤi7a7L@2h*nFpSYtx ĺ|:.KuyT D5 a4% d9#\:`[Cv[IӴ#K[7H- x}TN@' (P&Jı1A*VU8{Ċ"!!qjߡ}zky3:v ڱ֞ٙ]/%MFfnH`!==`,HߠJ;|]0`x=w ?2eD\0(+x['ј ,UL%JSa<32?$Q8]@4~w5XzxB׌Ɲ7Y7@2كR궎O7e_ icl{[FfIynfn;;E![0C2=ufK#m2p$Edqv*0[.aF829,Sd %#Z$ME3>F<`F|Ajߘgtʩt"i{,zg b 2rON9d̝џ }^͕~QiJ1C]H2)W[~HBUGI~1bRF}wvФtGu:NI =(Xܒ *23WۓݩW0bl^f~UO n:?EnDdpB   S A ?  2]9Xc=7u%G`!IfC紤TgNjl- :xڕ1K@{VFC\ ;.SMh@ A/O`] ypwwI~0ՈC!jkiI4r, sj,Z (ĸS0[蛻˧f<45&(6pl6Gg 6+"hInƃyC>oߨ˱ /9 \'55k]ZnY}Z$N@BGZ5Ԯ?}ۂku퀚"R_.yb?wW_[C]9Xc=7u%l-  x}TMo@'-niiD$'=D P;rujYG *Uzĩ8Wq7ofg5#r%Q?&5_#<V2%9{*b]jxJqW0}mCPunR5Dd6%ٖ%o MW+gw*1bqv9$~&ɚh6 Y`ڬZ(ɍ0Rۊq eM\qI>3;蟦lȖ¡IYM3x i" GL)(Ɏ0Jif+XsU1Q]ڝK(i1VlwM"e5X{5J.lwMBv)G8g8=88IPJyWwblt5ulQբUB̯5ZKX ЛV(ʴgbݐ"'bBuEe)ZsEL]π=T{zB^W Ld":QL/@Êu7V7%]e+@E@JL.)@v%_Op9iDdB / S A$?  21D0[uvXBAk rL dM`!D0[uvXBAk rL^ 402 xڕ/Qƿ{.!YD +k=&T"Ih6! a_RD~ӋDK 8wNVvo'{ϙ|=s2 @Nk 9$D@'"݁|e"*vJ7JQҶmP y%2s|77V'(F<#Q?ȩm$_\aC(9'pjʙߓTEeGUyS@]:Ua=eYCe풵R~y%j凐Ə>W#F1XY&ab+bKxe m'[ | Xek (S!M It5j((QYYbzfGu].FEu݃%MYvDz.Qg}aYDU}ĺ²Du7[u6XWqu.μƲ dv,kJڻrl^gM:`Kn1[|EXg+⑯ ^?OS^nTd1ʝCXɞ?8eHXf 402 x}VoT;i҄MB !Ѓ^lm篵%r z戄^;WE%RTp)z;8"q 󞿓K}37ͼY~v,, pp0np0Ipg0_q|Ƹg yIG$Lb__ ֵpOa n @9]"~Vf 1` ÁrSX* /Bw;wA.ȸ W3+xނkHR呰η]B#:@w~NPBCI;c+`b$z s13boԫ@g6*)+_X" mEjՔׅEpc[kW YB-zR*隇g^.sv:j3ʜ{Ş yO-{ե~Y*b|\F*^FfƖGI$DOlZGe mqcM%n'b-&* rDF8p'@8UckQ)ى,&5l$KX0i۾d8mt.#&k$FAiÆ&lœ-d7&ɞ;r (2Gz^'9vG$K1 AyvDc~)Ef7#~@€LRD9'y$W)0E䧎ƶ( aG(%G^NqJkó^CT]BUkPIN8,45]՛65ěF 0]; j1!m:֢\wbXzl]4Ơ&d"LS$1Qٴr5LgSUe^סKG2rPie%Rk#XEfGzctPS唖cEz)MBJԪɺm(.6ߢ`)o5ڔga `amiT)0q#F@=L`ɥcNcST4Y fMJѤj.hDDwӽ6^ a SW<[/D˧x"xM 0/(vrx*~}mtX2M QX+6*1.;Ü1oUكOلb sJ!=R%'m.SKB..J<룱OOT7T N*ϥ&W^\ǹ]\ \z7u1i>kE%UFQrM`45xiű(4X*K5ŧ 0bP71ůN{\4, PNӫ>rn]g/zlJ{vQaFFCCgБ&<ԭwA I/)xVoT;-U\Z!d=KHHrt^4!TTrOT)ȩʉKynR$,y3oo~3gW~3xW݃xAuFplx ~@㒲>P6q`DytBLqfށե8XEXr+>jam냝t F80i@}+\PXK JGC5QQ98VIE+fZK _uOkT{ 5P[>܆_ݪ[^u֪JBhfB4GH*ŵiԕ&*UԹR~h }s QIi&պvt9u9ukȕQ1z>j^g2Nu5~ ^vҚ@3OSq,܋c&9Mm[0m6OSn$BO0<Ǥ8ޏ)ӸK$y\Y$f;d|:IǓJLkes={q2[:.)Y>Oc#FJn9ЎxgJwDEq:hI@Zh+5>>hvPRlpǩ-ƊX-\T/d'_dpE- bymjsAXkዻ>4[ 3 1E3 :/1L~>ia@mev( pjv*':܀7ɈNqٸVm/,jvi?p@ӏ~4cG?~,g_GfjEd_B+q/*ق2^Ģ?M!S0~e3@DdpB  S A? 2U'!]o`!~U'!vl-  xcdd`` @c02Y,P1=F (`d?3%$ lʛad3|-wr^rrmZ> ,@RaҊ EL2 PӾM9 +ssd暄4F0T]&an 7b31n-ЭDĺ5͠Iwrmĺ| .OF[wlbsTbj453Pa<% dezFzu<[C3wnCFl- x}Tn@'-?niiQOBH;Nzʡ9:#ISJExx n'H];N*!6Z{ofwĨLhtr%_o~H&5_<V2%iJ[PA2Ǫ'car*@YV+!aޛ^;4y.DQ1:>F,=<$/$Y3ִz^0dqZFj&7àKYl+Ʊ*};WG;l4u-\w$I/yVl?X8|5)iÀ^8.8`Z\AIv߽f덅} Cu\gEEIjfGn)޳Qt{#Se n"3M <±=GfEYWsXy[gŰDn`K75Dfkf5]K 2@oXMNԫ^\9yK,KTP7u-j},.5w0=s!hǘ*oq<~qrD''JV0[]#w]޺j3Qw佸>/Ǚ^DdB  S A? 2NO-+kB9o*bo`!"O-+kB9o-a, hxڕT=KA.DHCHc#!1b!j!Bb ARK+4Z6wiNYv {w,l Gh+x(E\!beJ)QʸKUn)11n Ythx+r )+B҄3@ڻQ LI\{ShF]MV#]b. 71IU8H.|E ms$;76W\lTkO:e)j*? dI)G/HC!%e[P vL-a, Tx}T͎F.{ VJ\2dvv3sC(Jp {v{Z=r{gg5pi5x8܀wT= V]_}Uu u W,;·٤k ]iGOX?Rgy5)Ҽ8uS__;JBB a{;sO1%o£.EH, qp ]f!a+[]GZöo}h=O!سZ }S}p`\zOd8_H;Qb[ˇvE<;\YVs;EIdMD2 3-Ya.p ܵm4:EG c|6A),ypT MfA0q1UޠHTdʪ,N)KV8r&Ź汘-07G=\4l\'iYuY$i2Qi)(/jӼ81/lTgh*_&sX}_o/0MpRL,5]BUQ^bEreRG*c P>Yd %F} IW2iyaKfIq^J=}>UaF0]ڴ>y˘vH<,]w PeF%0שܫcFl6[* _! l"$p 8T\C'ڦ[Zw]6j!w [<(MXNUw|џtgт]u֞a^mk^悙/{1ܶN3PH\Xsh1>yBoۍDd<.110   # A 2i5 Y"I71&kfEho`!=5 Y"I71&kf~3[ZxX~t xZkGzeK)-HQ6e ԥAK/>Pl I!c[`B.)[=]Sz?a]ےf4dҌcYowO;3ޥ$KxHǎ`Q8gўh<)(BH)v`.?~!:>[*{)og\A<s=B;hn(#e[KQ n5I 鰫8g#.B<8JԺAGsSzߋ8d22NK(gߓPTMs"/YXHkf56,հ0ꄴꨬYnH]ubV iSYE!˪,f56}1k+V̪BZϩz*S!LfCzWYyu6A4Wf;4oP;#?-tK%4cŮ׎9c#ҙ<& y{Nu设<]Rʂwkwv>*\yij<سL2H:ePʊuG^Ye.sXe+UWAegmeaXgmeXg=cv/G~zVggu,-vp#ՕluzWg%A1g91zzƧc:V.TǞQPRǪ6v[Ǟz|;t0a?.iE%xcU+y}dȜ{3ój9.kX弊I6%ˎގJIAł=_[8ynղm¹s TX@-<@uf-? ]n\\e[Rĺ¹sK*usTusTPkt 2-]_b]n\%7;X[8ynIO&¹sK*r 2-|:'.p.ܒ@-<&¹sK*wCr 2-usTo4sTYX[8x: ?ߡ/vtaoѨ98Dd<.110 4 # A*2C^Rg@[ xn`!C^Rg@[ xB7[ZxX~tVxZkG!i4&4$D9XU'J)>C(PHQ?Ci}) SoB6%=(YcHJ3g~3owfإ$C)B4;E ?}v1;FYyJ #dD@D*gTv6Ԏ]:<xp,Ӟvʺ.2<j Xՙ~vXdts6r,ăYUZsBj ͌дz.&2t)Az4Uu(/4%toFhqYwG36:#Mls ak{3]}^}]osrXqoxzʓ8=KEƕBg)*TV;ج0̒;QVݙUXuҮQ NGYtGY9rހQV&+3k^f ^-5#5o_#?W7}ZbWpOJDٱݨˑ:T)cHuHG׬QӦo)v0fj4nE[MN8yϛCϖ!^^dѽuA=ŞV1ۓ\~9gO\)?ȶ9 ¹s T=w@-<@}*9s.9?s 2-P}y2.p.kFR2.p.Հfr 2-մ@S[8ynI e[RFp.ܒ 'ĺ¹sK*ĺ¹sK*w4sT/.p.ܒ[=@-<@K[8ynI.ǚ¹sK*?e[R9 -<=@-9tπjݤjO?~`0eeȬYkV@,fa^.{9ìJ>̪1k #jm2bֆ Ofmx,YY"fզZYi-t$vK*$zѢ6i1ߠnFٽ.9tK!o5[{g& vHV@_‰%=gkK*_9 qaFcU3}2ggz= .kZ^:McqMtV^ˎO书z kZP=}WeNO@ROMsM?={4c9cpDG3<$)prGT+EշEZ|qm qj4p=e#d {e Lq>B;ʲ<ٳ@PV^,zl`;z``;ɺo`;ɺ`;ɺXB̂m5k8XO]Jl`5\8v E޿૥2}g"\ϯгt;gr:v-cXu|d c#tW BG]g[)va_1hڊJ7ܿ?&'W?n<1xz9籇`1>lޞ3U1ra&?o'sz`^+ı-K?@u˘vL¹s T?5 :p.kDFg-K?@5: ¹sK(js@-K?rxvX[8~n {X[8~n Bp._w9 VΥ[ByUsPm-K?rx *p.gr -Ī¹sK(d-K?rx*p.{X[8~n 3`_[8~n eUs3H9!?Gw(c/|u"|==> )|Dd8 B 6 S A,? 2&]YXS~Q7д {`!&]YXS~Q7д^ p/Fxڕ?LAƿCN̑` "+w@#b‹'\q@6z5R`ekabmC- XoaVxsjȭj: F!.JT'GEUwC4z‹j !!51:^7BE) <(m.p UtLRRCj)$ NE>4*?=JE]l5XMoCTTj|pZGBG:( *v_TFhҔ s)׸IRR(K^&ǼAc+<< ]< \|~+-L8r>-̩PFQYBw\WpUE mp׃듁vK=z6ԪRdP'Y$i;kjf \>Ru[kY\:ƭ5In[ZSo]ۜ}#y%i .1C./<20<(1\qs`KOeAectgM$La8> 9|= ӫr|~t.; ߝ߳V86zCoP1k*7!  p/(xVoDv> aݖ*!HUBڅ$뙱gm"!U@H&K6j BRzAAġRyoMCTKo)j>3Fp|o+KZzK˴;ڝQ͒IY"Kw&{Q &[\~ϸn֗?;&S@uAqW]o{+҂BatQzP9^el^ov]R%Wf{B\m-'%S{&Y^ر׀u:b"V ];3V}+el4k 2Xq)Y\]ha֑uC~RﰌGuwx:7wZ;mv4R9Ή!7ͦSyfCzrr_+fEquation.3y 1Ga()b a a a-1 e -)ab ,a>0FMicrosoft Equation 3.0DNQE Equation.3n.39qObjInfo]yEquation Native z_1074519029O`F;l;lOle }CompObj_a~fObjInfobEquation Native _1074855010reF;l;l G36()G1()G35()1-p() 34 dp c 1 c 2 FMicrosoft Equation 3.0DNQE Equation.3Ole CompObjdfRObjInfogEquation Native  r=aa+b,r + =a+1a+b+1,t= r(r + -r) FMicrosoft Equation 3.0DNQE Equation.3_1074855161TjF;l;lOle CompObjikRObjInfolEquation Native G_1074855225oF0=;l0=;lOle CompObjnpR+ rz perc tFMicrosoft Equation 3.0DNQE Equation.3z z 90 =1.65,z 95 =1.96z 98ObjInfoqEquation Native _1074774235|wtFù;lù;lOle  =2.33FMicrosoft Equation 3.0DNQE Equation.3J x 1 16,x 2 16CompObjsuRObjInfovEquation Native f_1074774334yFù;lù;lOle CompObjxzRObjInfo{Equation Native FMicrosoft Equation 3.0DNQE Equation.3 16!16!x 1 !x 2 !16-x 1 ()!16-x 2 !()()p 1  x 1 p 2  x 2 1-p 1 () 16-x 1 1-p 2 () 16-x 2FMicrosoft Equation 3.0DNQE Equation.33 G20()G1()G19()G2._1074772457~Fù;lù;lOle CompObj}RObjInfoEquation Native O_1074857455mFù;lù;lOle CompObjR33()G2.2()G0.13()1-p 1 () 18 p 21.2 1-p 2 () -0.87FMicrosoft Equation 3.0DNQE Equation.3+2?s-;.\\oI(3fIa=[džIO V6V5p5iW6ԺE @ |H)V ,ilA&9&Rv nګpFҳEQP.wK^7( c QjP.Gʀڢvze0SHU6 0Ii,Z;n`.Ru!E[t=puQ^D<zEha|Dq ː b>ꚏzp>C/Ls_cWg3(DdB   S A ? 2wԫ>0d0n`!fugf 8)N1 $h xڍMKA3j%oYXPDE]nB]:Axbe/tnJy萗:l]Etg`ggoffc45;ј݉*[W)7JQ{Ci˲Hfb0h7fG7IGK[LPqU}~ ? a"6l~\<;Cfr..V jTQ &\R%R#^l^jI]q_40 0C@Nvcmjfmؑ4v0d01 $h `xUMoT$M$m:tUWs>6AHE%Y"!92c$CZAeb5H,HY@/`#pdR@<3s޹=w\=zi@9d%?@~>N*]7X4cy_x<q6J88+am H@kʆku"206qP0uɘןSSf<6-Q ݿ {P9xB71_M~XV9Vd!̐.ف#"2Os0%erS ʊ-yIn)/(PL׋ͧ?("D~dQ<`q2`zXPfмgwPyl\ZXkfQp&ϰc JMډ* {jƌT@\1XPe&kYͬ$$? ?4@}*y UC$?Fզ?4EJ.R\}H;>4*cܘ}sDF1FSU^ZAXjUW*T55RَF †k9RI~ \: ?\<&ЂrLSCd=>KxGeP_Ydy`/xJxC^Bs;9# 0xVKoD.{}$q^KVb[HHk$DBWǏʌmٞ<Оgr~ TDKt}WUm{ 4< jp@;.J=~V= ^?wHiŜC?@ &EU%i;rAplz؂g B']d߃ŽdmK6ᵾ \.TTi[>L+?ȦBj,9 R 4BLZbO2^lГR7<+㈉[Yۏtd}Ίl4bd"׀bnj5ENC?f%Oˑ_E̱e]ʠ:L#SRdsNrVF(I$D0k3[lGU vP g4˪Ȏ0HW\!Q4N[C& en,򳆭XR>p{Qq쏔:A7>㤈*+j4;IYL%TGgqoCk"%KbgB%* A.SdAP>&uZ$5*sv.iqCHKvL%iVyԨqPLXuWc05$LgvpoO vN\ kdwp,O.pNtƇFmAl\?!tdt4]kvunXtP1 6qv[i25ܿۘc&7|TKuxg:3ۦqSJ>MO͘7Ӳl.1]Ә t :i%ttML c ==PPKyG_Ȏ4%/5#=3>zW`E."\B=տ3ɓʧ~Ckc:!2?0N/(xJp,fDl7?l pwc;&㰖Z.["`Yqȶ+'-,7rƍ8&<+^)J>Jy4Jy<ڙ(^tTy% „>~c9{l2;h)\Y|gw/y+`}0wϗ|E:HfCc_˹Sz \'"C_Z'3+A;;x_S XwdA;;x_S X>U0EZuMɢJ-W}mu0EZ%R7LEVj񾊧ɰf(RU'AzhcWZ.<ϩ9z^% uxXx>ʲEl1`(YG怅,ϩ Xxfņ]UI̺]Vl#ߪ+\sf5чjg5=rGgF?W7DG\u<e:>߻wxpgu0cb.k6?0#U'A!^V[3Óp;ɸ,#Onm;)ϐo>Ux{6O*U]82W-xnYSB*h⹅k*h⹕(-UEcϭH4t"B[4VB]B2 UnXs UW$B[4VBս\7Bf-xn {Bf-xn=sZ!PU] !PUAuYtM3j  o`%xڕK1_j*A\\+MPAPpA*BDozz< Ka239e27$%@$+B*\0%V^3VE^[*,gPڶmTJe]n1j4:yˇ~]]d<&o]p󝾷}y"(#pP(I@I(c+ո8j`?r/P ?1Wb|,bXr K~ɒG!ug0]˰ 44k03NW{q]>.=#_k0IJZjezͲQ3a6zòZPcc`em,-6G2w,=u?jem걿7l{IJN=ΎfY؈~/&,˚j#;,kTXS{%XbNوbY#`b?-b2j,k#T -io3忧^v}mި'CnК_.:b^ o`%x}VKoU>cI"fc{KDBa dGcg񌙙*ETuk؀lT]ꆰ%"{<бsϹwιX Y;gx@v3 4{x^G 9>l7{ٍ"7H$G8f­K[č`@ځ9s%JO'py M"m@αrmN-x*ſ%i&y8)s\sE<kô4 X&+Ckqܯg'W`磀 /gPe󏌓r4x IIk^E,[>'Ͻ*|֕J J`-2/ S^d Z?㵯Z}>klW>OVl﹔ e:ܱT6h[Q4pS-ݣa%{lsؾ[L";};qSS+lM=gL3Ni&])F,sbR5MN=7 4$X0q;.,WFa,#Q/c'lwRVdD^/]6}ua&E˧ x7ڳTGw4JW( ^:P~@0S$V^L_fSoDᔰav#bqѮ3nx$" olOlu :;.d}g׌YG#¸sx9%;ƻ>3p_xYӺ}8}ϡ#4&ڠx4&UtYffv_7)eʅUðϳfrAC+w$@թQQRq9Ծ<3f9mKjafq^cKsVJC*/QnCt

5H{(+r\pb\:X2yh} LJydzR (lYVn0[-W [j aaN哨tC՞OKdKnJX(Q{bRdo哄 GG'Y6h o?j |DdPB ; S A1? 2gn /ͥ`!gn /<D- xcdd``cd``baV fx$\,F^equ02Ȓcjl@?H @fzM`d0B2sS;9y<\{=bF r؁"+#C?#i Gq@penR~شG<\BO܀3/Uvq^i"gpۅ2*\L$kIʕVY c0Ry#)i1Id1/avO!;P,ESçמUoY8_AlVy*LWZ/ԡ-,m oL| %~l۸o#O +ϝ]W厱a>dc"$EAǹI$I0J1/x $7 @!yWaM?IK4qVt$Ue^L.I3 %LE2ӌ% Y7s:c/4=/1 1MR݅ jJ`\H34Ee.4"3H; =$;jeSMJϙmY^qp `=[ p>LT&(Euՠ4ᅲ^ڤ;]%Y{nkeڽ힃LO;94CMx lwy*e.'Sힼ9=VotoT[L ՅP.d2?·0Dd|B < S A2? 2 8{E~v<6?I`!8{E~v<6?STFKxڍT=KA}(x*(h\Fl-"?,ٻ 0;37w%UԸ UݼQ<ӴXV Rx\}m&la+?I|~; 8(n7W!"PP~5ˡ*t v{?Ę Tr\rj@ƕͨ%n4Z|IN57bX͘Br46&#H5k5kw*My&`и+2ߩ'q$qA5gri?n3S :AM ;RO5mA;(%͔z HcV익ʪT2,cm[*ʼffAT=WEuDV&x:H QhرÎ"RED"SS})y/rY"KŶ}SAk60v' ڸabZȏ0a\jhy?FfaS< Zd o9CAF5 rXr78ia4,N%ILv0 ؆S,us̲n2.Ėj :NU0O3%r' &̘!.rTz=_`a'"0"]Dǚ|"0RQ呪.C%Dx"L[MY.Ce{*څc/y{y3=QGoJU~oF"nCk9LO-DzEjP]Y_UrTIUm \_FYA6GV}Hn`c%(]8,6!=B-nWH]k$(oR \955R[#U4pgӛ y~zuЦ`vZӴϧ4ޯzԯ6]j&h oKrlqpyeDdB = S A3? 2c7wmM]t?`!77wmM]t4- xڍRN@ݼcXY)G(("4|C( PPc|>!bOٵ<7;:A vaO:ԡlXvFyĦźԡQQv%ϸ\mi۫\ u+yGgմޥ4xz]<=cjAkC$LoFʾ<;14- x}Tn@;Z F=%jB⟸ !@U\שgy*TBxx n7AE:N"1;~ͬfw 7pC9j%֏0M ƺ\Iq`4e<& "NnS +F&{}kT}'ڶ6(owh-* TFC} R6]JYVT?x w]<9ޘ XN "Iab1r#d :;F!%Ms ߋ٪dz\^Ɛb]k<ؒEIg(XGM9'O@_*"u;8H,K5-RIҤό^e83dZM ɍxOhOs,ұy0gzY2?Kҋ\ϓ G*(ώX`[eyf/L/1 q 91B]HEQ"o6hHUd, y.5NNII;-My^hpNg-_J'u5-k̼814괆:KLuµ;N:x\56yUC _&Xn vMynUԮХۑ߹ 3!ߢ7'OO_ G~5W|*r_U?2Uo}jSW DdT$B > S A4? 2 8goR )g `! 8goR )g`: Z0= xڕOQƿ3QiDm!Xe (JuʘhԀ`/7n+V. +u.4D袞N $3ιsΝ *zA(3b{QBR!%*7r ؂^*͙ÌX/bmrυ^UcM$<8>JkZU&-ؙt9 \q'j1%~$+[Ca@{ D6 : Z0= RxWOhUf7I$4ڵVE^*®iVEPݝMlfI%+Q/= P`=xc/bA[A/ړٙeS30~y9t܆Yxb$| Ig >gA:";FaCȉ.pprP} &PS*+  W( @8JE˒7H/G[Iq)3OώesJ]b?9՟("2$̜ЩQoUDUVJVf.`bR?کSoUT> a+;[9)9V''$ʲ<2M\Iv~::yXT%qK/_CT1ʯ 򐔿  ⚥/FS'^orPhxMfhѪ쬷s%zvg 2Tn8o2V굶j~Ю fjýZ W[5*ڰm -j59Q^𙮙̱[lzaYµYqB篠`-m n_$|bN\VQw#fFfy;9Y;[>;(:oyUȵi}骴P] 0֛Ifd)U#/G͚G.57v32!ߩm3[7+XlEXkKowk>kr̭ jb02Hpt'Є"c0?p|R Ѱ!":na8w鶞A,2N.eۦpt.l[l=יmäLN_Diqoh6a k{nyt-`'Ȅ}3[uq͌To]7y0-Lۦ.8fsLlwfϱVZs C3@ Җpc48({$xh(84Y*CW˧˨8/yjÐdrGd稆0.Z+vzZ7pȢҟM-BƎ!Î!6,,}hhk1KL1QG'x:fOkrܑrapcq9~ riGsgY.aWBp<4cjlwCFjew'/Zlzeuci) _AJz7@Q>/Ri~(\KV§.s s鈐AEJOť3d\-VQ\99DlD1("b*҈Tjl;4mK'|)8A DddB   S A ?  2,?crT:o`!,?crT 2aK {xڕkP}mviRԑR+'m];M+ Tiѩ+\BpV7n\Zʝ..}1&9 $y7399 3L}D]srdٷ֨|Za3mѵ F9;XA jD%q$/\ 7 =nsSW6.Pכwd{UcY-ra:F ͙>?wnCυt5?"HǧWQVrsn"!Ο=Q-By2uTTcyhڃr:7y |TP> Mc(q>|J[eXQvM,LRXrTVe[UnsrVEajDge8;r}Qei!Ye:Le+J5רMduݯ5MPj8Wn'zhr#toЙz\EW:mRU^ߢ^/HW8zh~F_:gDcY \H >K.^vXUj,0¹ ,AWcBW`" a,ks><-I'mS0etxX ؈uNv|\5e?p14&ep2 8\B>`+ UvɃY]홡ʿ*+PIrʚC E0)jֲ) ?:smYkcKkck:t:QQm0[ZyWEjm^5Veoʕy 5ǗLP7 5iM!/,gl\_K>c6ȯ-вo4 öA?5pVBd˦-P[5ja~!LMl.g:Q3qs\fxUqnP2Vb'0Lǝ(vҐ6uoӃA2 ;G4 zeܥ0ƨ(> Ә GxI:CK2(+yVI际·a҂^xw;rѾ"L'ߍG2eCJO:CKG3OQ`vx|OF݈FqL) vOZSu52ϝΰt,lX}Ȯ"Dmw|$}m@F18@PzΡs$Nʜ jI?#D = ,lF]Xm}no6#ljT{K{eepw`K>?LiGH9 t2J*< "@󠰐'@,pj/y -! :ܧnޗMD',LdnL<KyT2EQgi7Kt9ց %l8ꗳ[;Q/9xVEvyJ& Bu->ԡC$9A%ϸ\li;|K =iͽKGi($xz"\iC$LoFʾ<;14- x}Tn@;Z F=%jB⟸ !@U\שgy*TBxx n7AE:N"1;~ͬfw 7pC9j%֏0M ƺ\Iq`4e<& "NnS +F&{}kT}'ڶ6(owh-* TFC} R6]JYVT?x w]<9ޘ XN "Iab1r#d :;F!%Ms ߋ٪dz\^Ɛb]k<ؒEIg(XGM9'O@_*"u;8H,K5-RIҤό^e83dZM ɍxOhOs,ұy0gzY2?Kҋ\ϓ G*(ώX`[eyf/L/1 q 91B]HEQ"o6hHUd, y.5NNII;-My^hpNg-_J'u5-k̼814괆:KLuµ;N:x\56yUC _&Xn vMynUԮХۑ߹ 3!ߢ7'OO_ G~5W|*r_U?2Uo}jSZ DdT$B A S A7? 2 upBGP r`! upBGP`: Z0= xڕKhQ3MMS*jѠI MMAj5U_EPTZ*4u#.V룾J7.ꮛԭ+W ۊ.d{Ls'je^9*JՆObzﭗv{byFPMUuaFk}!)X=s}[ ^>7g-5"mRzdlKG=Ns=2k\cסqpBB1l^BZiu'W(&jhPvUF-(,>R5#s|7JXɪgA:";FaCȉ.pprP} &PS*+  W( @8JE˒7H/G[Iq)3OώesJ]b?9՟("2$̜ЩQoUDUVJVf.`bR?کSoUT> a+;[9)9V''$ʲ<2M\Iv~::yXT%qK/_CT1ʯ 򐔿  ⚥/FS'^orPhxMfhѪ쬷s%zvg 2Tn8o2V굶j~Ю fjýZ W[5*ڰm -j59Q^𙮙̱[lzaYµYqB篠`-m n_$|bN\VQw#fFfy;9Y;[>;(:oyUȵi}骴P] 0֛Ifd)U#/G͚G.57v32!ߩm3[7+XlEXkKowk>kr̭ jb02Hpt'Є"c0?p|R Ѱ!":na8w鶞A,2N.eۦpt.l[l=יmäLN_Diqoh6a k{nyt-`'Ȅ}3[uq͌To]7y0-Lۦ.8fsLlwfϱVZs C3@ Җpc48({$xh(84Y*CW˧˨8/yjÐdrGd稆0.Z+vzZ7pȢҟM-BƎ!Î!6,,}hhk1KL1QG'x:fOkrܑrapcq9~ riGsgY.aWBp<4cjlwCFjew'/Zlzeuci) _AJz7@Q>/Ri~(\KV§.s s鈐AEJOť3d\-VQ\99DlD1("b*҈Tjl;4mK'|)8Dd<.11<  C A2^t? b-(:o`!2t? b-(Ѹ[ZNxBr&x}xUG{f&@S܊S8ݵ@).N  Zssr<ǣ'|Y{;sB!Nv5ف?d8Z_nryDr: rDjYsa@G/,y$qWӆЛu_?oe6,ȞyкSͱzZjzww?愈 |zX2$#G:9J[ BoJ;^z^W槽O9|QGAZ\NᚴL!x3+y5'+_Go/,M5"xC?Qq۾1/dJ.4|&Jj|^O{9awl'e6tƾk-5o=t'>{J?Jp?,CGZm^{q/5汆c EuHOWj5Uk8ZifR+9ZߩŜS-r5AMTj2&rt5ցr,?!r0|~,1D+WM|~&S=39dMdV8qTNsJ|:Ӡt%=jJ T114cP_99 b?ϣrn5.pQùƵorc4V=`w׉\ (W\LBtZ`-7hXv|b">I75envV~ %QSQp_1ՓK\NuE\Mu_?*+",j/Hzsr yzr"'BlbԜ5Rm>Q_k\6 ;^-<*V?`bm S uUeU5TUEoUUcj[?R h*<U9^J%x*ʳTa 8 Ź*uTYn<ګB*[W㟠[O}.r#y1OdNAMPX4J-~ϸ?a?v3G}g{'xmyOz@VIYC8kO(W_*ǫg^<,ng|5g, xSRgA9r~H!ۉL|/ۋb(-[r(/EY@D'klH- "b̐uF&[ +Ќ%TDnl(Ѝ$뀫"X>(" tH .زSI_+ʢ@ QIdQSeym;,"amen%eQG~'jlX̄ߢ"-|X";͏|EҰC?!O-p?b@ue~PC|h ,,Z9bo!r;ԛQsԜSXvQVQ(z+*|B+ -j#!oF,:`~]:CЎF"9USd*(D /]&%d|Iƅ^l'&,AXL{*-AiF;#yj<]u遴K-zL\ɑ3QG47 A@-X U_5WEe"%d-\&?-|O a?-$sB" '[A-NӬ8d΂9l"24NS>5N8ՓNl\0pkLqpZǎH;v48(\\*)3';c,N8kYI$I<9$SC ԐZPc%@kjG‰''d_mU YA( p'\$h-ǘ@kXё߉'< ("axㅿ|kbC'ۈ'x,~. Ɗ§01V =q~8h&Ci(M\P #p0C扝b)koNq_aT4Ƃg<>s8/'spO??91Ӟs9{5+} !;=|[r_C>/{YٕO|R6 Y @I01eFYV 7]CY/b- b+A:PO7z^؞ t-T=hɺ :;<ּ]vͲ-oX{ U]\. +8FC?bm6ߎ }rs zm:B"O(^XȽDe}lƼOE2͇L7pďFhMx Gx!.L`<>^ w"[j&% v3OD\ΣF7 y`c-A 1Řr8K)c! ȇO:z;ɍ*,pWK z ky'u§z`;=a9Ž#j6tnƋXhwoi'LȜSsqVU! /:MfY975Z.c6ZW1&Qw9QWveS7;#ʁu̩Lέ;"jZ"@M.c=-] y7 GO?)g1~hvqPe:]Ρ3t^[QK5ɥ?@](b):qJ 3ʎ<)ߣb-*_.P -OPf)(EOؘsbP&  sT˔_MNCB8csbBKG§vbc-?3tzEQv{׿xu׷=iT#|CF#2rќ/- v?2و#8YՀJ,n4%(Wl[hTg\ISt!H6IQ[2%w nخm;Ȣ3:StEWgB$p` ί4́"Wb3XZ,ubS>#~[zo0|F9}$`꜇zbl!k3PuF]w9!w<3BtC-gXLM*;S[>Yg qƙL;kho#4a~_uIBvCVg巰kFCrAT||#65q39ģ팆z@|T}U佉7ױ6Q9\A԰k _yt,|Jsmҹ~/+w>s%4sϑ.jFEdQ5M4s]3EG, {k*F&qbk*Y~b)@;Woa[h +o$q \X15F tM=]SD'jgkښ`kh zC<?sM\c~r׵Dr-\16Ww-DbkZ,~w~D4o7SF;O(~KEt5K"fo;Pk k}-|"^hqoH !G,ȳ <`w7Mno{iM罛xҵv@,Dr 99?YN*gsȕ ;}YL:y?T^O6<G%%7ѷ-,Zr\8Oc'1kڹOE]|kYp$KRdtP\AW> h _Tʀ@iPS1,PU & dm@!|*eeƀR2k@1 !>eIWj $lFe 'r9OPHVr'E\2rf)sԖjR, ^V|8T(TE~ޏ[̀4+R@aX ?m_i?i3CE6O<}ꆿo4y&zw9?SJyZ@1[ao/zQ @&ʂjR( RzDE +YΩDOP,uVPS7(K CxB3JS&c;uʭΣS)ڏ:vo;Q2_|ubh-} ׀^5'S?Gh6^OMjVSG{Xk/+Ɔ4k2Z*&ɘMR0J_P15GM=y,V*QzJ.? ~ F uVohۢ5Q uo!fiVqfcN *Z ܶۅ}X> Rg褺FQu RbN;nu]j1ۓ'P\eJ\5ۄ9mGt1qtyb:څkIN1/?|܋ <>sK5-5^fglp=?>i)4vv\Rmm  -&RjtЖS7m%mma>E~Ji1>h;j>5nRC4B#)Br(P\wJY@ߍv+0׃(~R)~R)~O"~ yП6ϵ#з[LRE:j""GT}'зAc3k@}Ko"+г AC㠶UJ_Π4L=m|bab$Ƌs1-x.?ŏt.F?-+o"CE?G2hV`oZ/=sh>v=ϥ }y}%]U}+^;أBt{%iomVb~bڎj;ߠh闡q.c2g,3GObo?/=h;jڄ}~-< ͸Np((q+)(b[(*Sc%0fRB 'f~6`,"XH5`,rj*.*kl_Hu6K jnEשB;Ҹ T۸FK1*f"XFTԘY1kZvzT8:.Ru22+257. A=mȿEC8n=Wr`36_+sep< ?W>ҹ%\;ϕDwSDs ٰQds20'Q< 9R)92)rk))y @yͫӼmb 6AE5P$cayb')y/yRh fQVg1Sfs%e09s7 ):h*``[AWv-O?&hy&E 5kk?sAgs&EXFb` gSms*dNL` 7Wk4 6EC .XXvYh[]=wi&?G끭^:4i1^چ7wo@ij>N?%{;Y;w7_Qwljٱ>`F{^!+<:{ɾy67W-{wN> =|R`پyfr=S}; #Z Ĝw@wUюAQ-讋9>}^'!o}}>=#:foxy{x;!)?/io]~윷K'>˿K+'${b$KFnpA!})vb;!v(1 +9/Xq!P`\" Gb4qOxXh(s="rW]U<D'':Ȣ!ڋ@" %i0^tH%Z _ 0ǔh3";s@g EkĴBL+ m3nu::S }E7Oܷ!_}_QphxG+GSNqyM57ɧ8IN*~8>x?$?N{?@N:7G@~U jm@FE6ͶUVmmr:V yr]-7rk2a[ZgpϯQîlpM-1>;7:7@z^k\ \z/{'2$ø^ne6ٶUrya߹9O wP+Oiav---7Or\+/WKmsm+z^.c/wS屹 ^nz6ܶ\V}+#ӽªsT _ͣ6޶+rl.mrt/{F͝@zqo[r%6Ͷ9b9a[nq0n~ږz^~jؖx^k͹mϵ0тmnm\r#6V'dOV;dPmOm[nN}i[n6}r}3lnmy&ۖylmJٖsl.m 3)rm͝-7o3ŶFAk[nnibFab[nNdHf2 Ӌzqm[nm6WȶܜQ/esm6VhelmyVfۖ\WrscU-7*\Zrsl͹l͏T%`[nާ*jr{! VV rTzf[n:xz^Jgsl+Sz <5}>{":</˵rU6WܶL{^M- ¸^n^gz/WrYaw|Nӽªs³M-<{^\ =E-_1075124924FѼ;lѼ;lOle CompObjREquation.3  24  0.7 2p  e -38.1-m*() 2 )2(0.049)24FMicrosoft Equation 3.0DNQE ObjInfoEquation Native _1075106825Fd;ld;lOle ;Bzb  ) 1}v( هҵr}tﺵ{\:.yU'iƵ {>С6DOIr ϝlzJG+߁{[GZk:[[}nl#/D}hezFԫzm~m[@{upmUsQμTYh5g@ sv6vv wxxhZBXh{B^A9O@^{@DM!P˞v:)g:텎#v4YFoRiDYo 8FL4g(FQ0>^*j+iHG1H5ͧs-F%|؀BNy(yrcd {-+ :,~(gHOhgs`k|giTYD˓6pMF)>4Iu;;ޢxa Y? qrµv1q W 1)`)>zci^#$>A A`g5,׿|D{Dzn~IFo!鷒C;H;!&EڍWd6O;Hd}N{Șݛl:6Myl6X`,W/>LvNJ[*:r3lfmpDA𒴣tܐv7'rHz6w9wyUJH;0/v>L[ۙ>x>Y_YMژ XOƠ"mL'H_<ɻI b|B-+#-5 ٚ,L/;O~}N.)$_y;\Kd߸)=^:%Z>\C$':v#Si#=磻_`Q}fôo,cx o$-7n#ssg}c)lHN}c6\a2iAٓc9 &}p*IrA` o_1gʗEs|7rx~f˯rmMsZ*K/F=D6.SŒSYffocU3I+fV5 rUmszU>.3Fsd.?0SJ3 m:^3˯[q+U,Ūd1frQk番JhT)LP-9do hǫ0wʛ*?5WCki6!wЏU‿0+i&#yjefQBE}LuUbP=O6@m <3l c(a{/NYjnX 1 ; e{2xĘ:U{w.nົS|osfyB$ ?mh_@yw*dw*< گT=@c<PmQH Fƀz)>x<Sq Py$z.5<ƙgh3O 7ϩ|uyQ\,t1/n/QQWWs[3Yu,w2σ<ꆲ?<2O!Pr  !ߓq_9RSuI/aEQͰkу|=>W5{GDy@(|=}qCUPOB:W_y-(os[h?iwǽe26GfcRHT25q!=tz9ddb&sBːdxb32piՔ4~Ӝjt=j&S䈗S=xdxSŒ 5;:XNMNf􁢎Or2'y,'߷s]U#ͦrwA{TzFT0oY!WBnJq*YokZETB>IleP-#Y,Fg,~=,33$W,Ls3*~l{Ie=_flgqlUt6GueAA1Xi?dvO-w& =CKzXopKϐ^Z}o]6wfD/:aG^جg+cb~Dw6D,VY(gpBR}K,~H,CW=*>z |~@ "rSrՁ|9ճ@IH+R?-6yԽzΗa"_!G_ܫ~ 3x&w hmvO2͐~ <ΓuS~i쮟'gew~QCQ+xuxC*gQsdGyEz̮@y(l /?x֠"BBYO^"d_"OՋ`|M37y6>CEoUo uy#?h=9B/CyVȁ \7/!b.ƛ''jy^|+9SЯӟg_W_˟otYN[8C> 0@}vg,7_z-q +OPOQu:'wOIB~=pނlCzS*5 |[?F[x|sDݥ^uW-]NTifJ+d vG|0JZzKQ$Jsb߅9'pJu9!>zzP}׋DA!9\oJ{_Iư7>;7;7f'~_T}=d!}<Ҫ:҆yrH6,f{,^vVMq\1MO;: z1Kv$:,j1h4KF?;=lܒ;H$=b j[жeU<~+2T`eކ#!}ez;G$`,UFWln=+`'ugQ)u^3>++26%gvFɐZqdw]YQAϼۙ=fz|k_,¢oPs}8--eۊ u;q ѭQvK]3'AݳbvQnmq=3u#@S1#% %g}k< ڽ x4o ܷD_1964!8pz ų<ӑ{}?ۋԝ'f|NuKw7').=PN7q.D Chb>VJH^LťH{!x?JVEb ʺZ'&qq 9E ׃s?m|+~}I7+HO%5_jߝoaf fUVvuFznNM|מ%G]߻ hڞTe{3_=3sOI̳4o/{EAc(oEe Q"("I IADXb\0UWt+b nE[#EyQB\!m][{k0i4g CQ9RRN?ͼ5Hyʽ5H?R4/}bzk0$e561o 8ǖvoiy{#,=škH&aG[ϲo5ff_:?[zlef^ z1Tֹi-&I(Fb: Ęe֏2Uc3$/˼,W]-ٷeg-|$bFagC' Q9@Lp?:p;49M!`8o'uõ.[LH':Lj'WF.a йZL4g8͜Ishg:D;g(szsNvR4mFDqsm gS oXOHa ĩN{\?[w3< [6, Eh - <3F4+!(PP߈oHae(k\^s mu8ϹG+:.rݜ3Ovfi3N p⸷tŵ DwNG4r F D߇9]vaxn8g*7:7fy( <= ǣx $<KAz]*bxjsu{%9zs2Nl[/)-!6u`o/=UPݠB|\C|&&O{, ]_ޥg mҥ<߇ړ{q|y4bM>&.M+♡)JW\nӗZmQ̺Mvmck2ģU ҙiוEStNQuw|4^j]e(<[;ʺll %GqSm]e^JʺmQ MRUeTLX̤ Co!3VXXtH,/KRz27Ճl=|҃ IIQ&O&IL2 Tz~B~Rziꎶc(nRSUVrnJ㑌Yk=XEz<F%r=X鱦oqw'ݕz{~ X=Jׁvkݣg S@qO5q3zYor-n/?r{Wnm j[<$wݳq~mNG-߸-Kޭmt"xM ϣ:?w*w~=o T;4Ƿ{~}H.qĻ/_vB<^ZPTw2duCq/F{.֟jB;Owꛎ8:@>rEOyFuG "W=>uϸ/?_qܳV]w]{k5[8m4#mw9x r kq75j~T/ zVOML!KR\5Albg1ذoibo+J.. rfaѐr:E*%>VNJk@ODYuZָ4C_n%6,b([Ӕ{Q!)ž,2צѸ1̴ŭK"QE)G"шE٤1>5D91MjSL`)3WJIkѧ<1{A#:3>5\6iT>nG=IJQK!(hx+ƨ/:c6_:cӵfQ5#]R֌K‘v$%+I4kZ_*V%IxDWiE]h;bqӶ5HdZŌ! HV^;Y5ƨn'&2*;[<1v4{rB;٢-%gc9CNwC,0mTr{4S4;"־cI12әDu#,u&EZ"i\2>\C Gu&C2]JmCuwdqAh Loԙ,v"әl0*Yά$$}ٮS󓔞bhT؜B)g1rmYJjm/H4 gvoځYplo(t7o; t7J›{3@q|n-pC-]e;={]98A]ɺ7S_Sx~ [Bv|}&xO@WsAk GwVlutgxok$P.:XzkAp.6"R꓅֧tsQG%e"mG{})9m}mDn[ [ ںzWj\*7%rz\VRMjT/+^U?T R~B*?{%zE~ ڠ.K^=LnUO|\Qj7!븶R=%Pȷ"~!Ex>[-SAt/}(_Ajn}pMt2OSVq )SwE=AUW@=QKUKPcG\XGԩjj< /W,mխ 5W*V?> 2NT@p|%ƫijA]kL\ O܌gx;b0e1vL5Ρ~=RX\@ڴq-PcN :T<|oƌe|3_;k<9i5+ǿP|B"2Q~3u+LIcg‘gskgH 捚QMQ*R$U1ܗ.,nbVEO֪Ȭ+Y , !KCF=a:3垱FbXzƲ\̃j@=Y>-Fbr[TG=aRR^cJ9?kͬ-Y.dH҇ qL`'kY&I6773t")];ndL eJ6arXY<ٚvrow*ۇ »&)[ӼE =r0Nʄ,#SBVg+巾V*ǯT=:T:!@?[ q:/V'~7W۪LTFЯA/ׯ/oc^ǃ7SGMQ~#;HꟀPwUǁ_s;~x'eCp:WҿLiRվ [p_L=ǻ?6xD/QYR< yNM2N\]&8Yg\{$_A[QBS3_dk\T,EuBsپ`E9\& s"rljnNfA-dLi^jBM⡹bډ'BSxy67WJ!0TID;"M], alJfR6&Wr&M>XNlfW{/'>){ {g؞B?fU$ij%㳦e;0;t]2XbXX? Po#0(نq!:X VrjLND5As9.h- 䘠2`:&'埂q`L7 W;@p~aAyK04T F| 3AӃJ9%:Aq^Nɠi܈_H!V< rpPE~92[^)/OPLY($ u8Dv #Ȟ_J^|ǧ([+{gVM;K~YˁYhYQYYJ^l}'G!wI6x!{J .Wp .1n?0ZvV}ܮ>fV:ӤMbOVd2-Cd3Beҁd(dd{WVpw2EE:d`/G"탪4($d YV%}f0O'(9j#xJR^Ήk,u.鹮KzIzs)* gd^b/))28Z^gzRWDc?2IB*}$=iVӦ8Υ㍓j"7v1{%+q<.g\DK}גxfƴlH|PC[C~^}ݽy찉iekIlO攰l~b#Xef _i? <&k_#Wҵ/AB} %&{|܌J鵯_"{!Gz־)M^r,-ܖm.cugw[Wڸ7vn -[zWJ_r^"݋eh绿H_" 2m$QİtLp[[f!osckL*J&PhW2iP@BŽa1@|CphX#mU/Qup[RF?gJ[GZڎpV[ڈ U-rz]FWdhzM7e!e7@Yj<:Ψ-Qj4SfOePMC ץwzODY})Է^W?*h#liGK(2ʿr|P"\+w9@VKP~ W3DM93]M33/ųEhgz|$ԋ2<J |/>jtC\gV: oG録򠰮g2Y=~Zvj{9a\e h"1hB{uʂ %1\[Dʤ]뤱k>i.^?OR9Yϭk9N+Sfu%^xoz {yO?:{1&ހEQ:_q;\SAq15?Gqqi5a=-ҋ_1Y^'}s>29Y ^wC]{8U[ y-ȧHX>[ŰDǜh嚟?P}(5|w 4\AH#ʂy՞CvT/幧$S)W\3/:|>衄#׍Ei?|5'j~>΋=rHQ'5_|JF4z`i`,oC~'\LN fC陞L5,?k Q^@?N4v='S2 2WA.5.4)SI}1gRgdR6jixlnD.nϘb&w0Y*,N; }t+o/7z28i[@?u 9BA>Kb.*z1%DY.FZ< =2JNtiG/2ϒW`V^83t1 ~/B5h/2Xų$_4@>#dZ̠H#Nr ?Mt7=z&_F)$&0xU/10I#)~K^`Lqt0*IÎDcc>M)g}T~jL^7äh6>G+758<Һ@JjjЦId}3&SVDziDxq,N7>H#K5 q.ɚ[yg Wd|gS2td_b&]޽HトM=Tp|PWAdRLe!:tҦ[fw! vM{16 !C6?0U\R-3T0<$3Ӥg򗞁ddp}A=(3 奏k$3>:2^#}vuc{4ɭ|o=(.FXifH{tIo{ D{LA4{꜉pLF=eJc?*Cv2c/{E2^ seN xc>{ ʍ2}q.;KkiaO(u4~WHP:  B)ϗ2.2ɾqOu#x.Bqv.;O#}4DF(n~i(Pk%n!z5sq_@+=}tG\o;G؅2̾VJX&Em>3)!lm+3KA_ʠmdT) 7,k&נ gIiypf<~ޜk}D՛k}=?ܶ [}e֜/HJkB*cuf%qtd3~$ڦ[E5[z] {n>.EN6h-칼gž#*fk!uB?uSHfh?q_a~>ue.MǪ0Z.:8A` ȴe5zpKղ/7‘,yHW: *D[EPTC? h&0V4M05Ջ2P|Ld^"+ՠV{^Wމ^+Z#z+f%Dn;ehP\7wܾh簍EEoGvy)ҷ(^!kSF|- Ե o.9:Kj.qqln)Г3W]N,)t9?CR?9:,4O;'Et7kC}af6탒㫳Y{$ƊbJEl;=$i,KmԜOg%9lPOYnp|23lP-;ɜ&ah2+*7'u?kLT İX,f1 ؿW#d}nQ:eX33Ǔ^o+LÞeV67-A3zRl~RoS Vq6»OVxw|1L/3#~vό,ω<ZZY_y¿8;R qYDxdh,G n<}9/ kDmtQ Fpr5p)MFE(̿$Bx!/M ( /EDƃ& n=oo_+"h诸_ȳů *l3lc !\^/"\*>/B_]įruQ.ԍkGCRm 쏊d;BƫJ[Qvjk=(BSZ.sz.}iA :/{:zAe}*uY8{%߬Eb=ܿi=j=b zyzyڈ-Vg^ u]({ٌ]ܯ9PFI}qZ|cvvﭧ{Ե m73ΌUKUqqnEQh0 >uGKRT|#`ABP^L&$d2 @xTOptYg=sߝ:ceI *}xVY4V5[<Yv\˘%dU̴d- =aUHvVn\CѽOo.="|WO^ ?yhw^WD]]C^~=gF]XJGÿgW^uFygv|/|o oW |eUHFSIk]j.(ϲ^vi{̊#^{}e.ui%W*܅޵5}A{ϗhkG9BW0hJ #jvO Z؟rkuJv kh\'/=Oofu`wZƳJя`cgD}i&#D<u,ڳu.+GYpaNHE;'|էUu׽TʱmOi?VmݟqQ;NZ 'Ƹҥttg.tGZtn̫xg tSZS:y-M:}*/Vғ zi5hܬ2+FKWK稤'v[_Y WK#N鼃0.0!gIWEKHm!K眥qk*=C9򉴊p1ٿ~N6ҝ IJlŻCie&ܩ~^6t.T̑*b|L+Hˈ 3q10m緝+Ȉvᵺ,J1tq4.>H)$K~¾,>aB:VA Pl{~}6Ӓ->HZEZ9KZyou)](/kﴍw2\`cfX'J+tXˌ/י/Yi.b8suK~ $2!tK@ç~}_r%O_r#_?r N C+a0@jF(710\HޗȽ:̗z\f}4UX{ VJxdmxԿǼ_z6dodɰa|$Q7Dg~:?}ч<} K~OH_/ o{s=ǹ|=rrCs0T;_8Wk޹~mopc?\ kal -o}NwⓅ_Ygˠq?+/@ǃ<7\r'\nəG^=//W1OPy0T:`k#W؏2O߲@.au2`3㙰mogD' ]fnvdm0^Qe^>=E9_9#<9:HK7Dr$'\7ɛD$B_7HOq|&Ȼ4N6ї=4^(bPI]5̫e~:Ei``|RB`%3? %_ '#< O|@$HB0~Gރ3v+j\hF8hSi888 [O;?,Ȇ {auρE'=7yσq?5US_̯@'^r+÷ 2r\AYNC䯠Jj#[G#'%}#z3k3=f_!~lg}Y ~r?ļ\A/ ׋,^ZJ慙B'^ ׏~ _-#w!O%}U_5om=֢ˮ]՟"̳ms}Y_bRvQYzx3%K_NI^_ fLEO/la>Y?/bbۧW+E_22,#2ׁKKg{\vQ_D 7nlx_1 ̢n&3$3tt?ᛆ9ȓFL)M'TO'g}ͤY m=&ֻ't}Q['m?NAkiljmCgNgp1Qםp7{{/ ^¿VՌ ԇ 2t_O ~>|}#\%+!r_FAW*0VCM"1q8a'w'm?[vbOݩA={{ƿ7YzkUWBA2?N z>t/§B| /$G!y UD"rG裔~0ToԴqV ;O_8RY UPmpB< .q3y>q6{>8'O> W@N/y ]HbW A A8UPcnszCn'g#sOYM;Obl8n@PDz?/C'^9!C ?L2PǼ"S^>ύ79q+|y7/"їPFn=(p$g]3Qx(v< cQHiF1(uEa, m_6-ҫA k#uF+f<I}mlA/ߌf6  2ꊩ/dxC׍\ro;]uFo}֑#5U{sLVXl:Fy{cUs85W65TװIb4hVp|TS׈25ja:ʿs:@cѨ14w8񅺆qB%hR͡s]5ule5QѪaPѯ4/*k,_T}1vsssUsĻyʟ/=y9s,+( s|FN}_:^lZjG5Վ)kx4c)YQYAMs_Oך5^՚5vi'5hםZFw팲<휲F֪qA;|3e+/5k5NhVKe^WK,[4ep\߄8r\gqPj/8+ξ4Ҝaeeh(>ѨuLT>NvtFxDYcnh _h|avvLPH_F^گMu>e.1̵WYy^5={}DdB  S A? 2پ=qõf8[o`!پ=qõf8bnFxڕKPϹm>$" UkVR[T B'`qt"GAU\tu'B<7M_7/0|>"E1G@FX5M֑k)ڑbS4MASHC-̦h).XՃs+bmIDun:$XC2J)-*T);:p3@aXPFf"vęI=ƊKP:m' a{o+4|peX/ӴhNR8Sܖ$SmيTvpsϷ&T 99<եErs-6۴*Qo-%g1ZGT% }]Xe q6ZO:+;12;{7⣋3% 9\^AA$:H_o! 0h Sg/JDdThLB  S A? 2_؀"djuOY}1;,fo`!3؀"djuOY}1K|x]@ 7# 3P0< (D !5 QAf|E @1uH:aI  /D02EcnWPw pFYl)&x2ů7XW y,.E *K%c:Hݿk/Z-CX|3/y^8TK|$x}SMo1n@UH3zj%I "!r m^*Cܠ'0,ɅyȎXMmZ~,aOG=ʞctXs<ީi'=0.E!ˍҎSbJ#+<NAkvyh [;6IC}=BKd Q%ȱ{](L߀cK~ao;P">+bmIDun:$XC2J)-*T);:p3@aXPFf"vęI=ƊKP:m' a{o+4|peX/ӴhNR8Sܖ$SmيTvpsϷ&T 99<եErs-6۴*Qo-%g1ZGT% }]Xe q6ZO:+;12;{7⣋3% 9\^AA$:H_o! 0h Sg/J2Dd4B  S A? 2uH9sd.{xx!jo`!puH9sd.{x)#pxUP;0 }6W@#s$`a1`2\;p.0'M#5R$:PzBЫ,9j C\ӼG|wZٳ+"^TTuYdt`_6<;.<]An}ރ^<(աsxpлwOק:ZJ4L10,S;P1#"Q/Pq6 l#oKY^sq]u)GI`Ca=sdQu)y%r:6aoDJ䕉qb=,JSh&Bx;heDG/ exȉi>Itb8kaFI֒tF.m>y5ɏӵi]rS*MeuC]WŹJ꺨CӜPkLџ UYm3jԡP(6V-&j7$ -4;9spϕ9ٲ5}M}d31ZO&jf9[N'SWǻibӸ墦k&ePˬ1[\ Uрdcx}9ߊ1K,JRv)鷠8Z^DX"%xqZЩSPWq-ťr ͓|{Y/} GSO*A6l^zr7zu.tzpӣDs)|=* oDd|MB  S A? 2͈ .`{IGJSoo`!͈ .`{IGJG 0xcdd``~$d@`,&aw0*(e112BYpL $ lʛad3|-wr2n{~-Č`i iEVFA& DT#%"L@ 0Ňa3@\L` l.f*6LPS-Tv <[Pe[f0>C0>%ߋcڪ,LG 0x}TKo@;)FM) zjDqCHC/Coȵשgmy7iJ$T!+A2뤍hy| (8ugx2:Mff 8(/.*ӝrfi$U˗i"ds<5s!S4ʡ?T闘&X䅣'%7%b;Jf,P*aHMMV1`RΘ=&&Eq9DRit^x{tb켖aFxI;.! Ĝ~o,D |r &zsn&1Xݚ Zz{މыW``٪5O'v|޷|*?+z_=Dd|MB  S A? 2`McR'f}jVso`!`McR'f}jVG 0xcdd``~$d@`,&aw0*(e112BYpL $ lʛad3|-wrYl=bF r؁"+#C &F Kð Malf&0@`3lqb*T&c Az@;,%ߋcڪ,LG 0x}TKo@;)FM) zjDqCHC/Coȵשgmy7iJ$T!+A2뤍hy| (8ugx2:Mff 8(/.*ӝrfi$U˗i"ds<5s!S4ʡ?T闘&X䅣'%7%b;Jf,P*aHMMV1`RΘ=&&Eq9DRit^x{tb켖aFxI;.! Ĝ~o,D |r &zsn&1Xݚ Zz{މыW``٪5O'v|޷|*?+z_=DdT B  S A? 2 -o{_^xo`!-o{_^:>x xڍTI/A}U3c3a!K"$.BHpl?Ab9ph_u]iPtիW@1PZ'FoEEsݢ#)G!!6qj3;*>jOZx=6sk{x;#O^M JmBcv6M#i**5HQFj%VkӨ<5WS::AziuJh4@(M+Qɜ?̭lu> wENh]bYa?{ Hq\E ~ts~r@r.tuWŸ:1ά۝o] XôΘiLb9XרЦx xVKoT>4Iƥ 5R5nɌcR% ,UNؙ P E7] vПpg{y;># aU=Z#ZPBR~iF =/\kKbwMYٽ-eIZht:Yp/F94h A(|sUrb(w ށƑTh,py/ظЂ[*J@_PbE<닟=nepF%9e]xoRl, s ж%:'Q_~CZMDLkj (}]C1KkA-a|gYR <79ʍRb:zΌ:N0AF>N?)x٘h_D~CQtY0Td9B0q FQfȄpo@LIҶ.-n ciI:FLi73"SkgNb0**$#tjZ"lu* $l7ҶQ^dH+i]fG/70}4=6kzO У+ZHQFY[4e2x)0=E)[VL"]$ #=NhspCvI٤Ua&t,c>n2Mt͠]dҝW9v t[$WΌ$yI7I?6;-v48>N2Bjn=J1*0sNrl}X+`(ib.&T^^/ R`ra@!wă4aߵ< i&dJ.tGo~ {)ƦֲÀZh-CGKK-׬6]il_G˧`wBѶMVpЧ!qld5l59ML:\ш:h}jZmv=|IVn~M?l̀&! PgNl.r,-Y7q{9FCompObjRObjInfoEquation Native F_1075124853FW;lW;lEquation.3* )0.7 24 FMicrosoft Equation 3.0DNQE Equation.3Ole CompObjRObjInfoEquation Native  1 2p  h 0 e -m-m 0 () 2 2h 0  2FMicrosoft Equation 3.0DNQE Equation.3 2x_1075124478FP޿;lP޿;lOle CompObjRObjInfoEquation Native +_1075124252FP޿;lP޿;lOle CompObjRFMicrosoft Equation 3.0DNQE Equation.3 m 1 =)1h 0  2 )1h 0  2 +)nh 2 m 0 +)nh 2 )1h 0 ObjInfoEquation Native 8_1075124539FP޿;lP޿;lOle  2 +)nh 2 2xFMicrosoft Equation 3.0DNQE Equation.3o h 1 =1)1h 0  2 +)nh 2CompObjRObjInfoEquation Native _1075139912Fd;ld;lOle CompObjRObjInfoEquation Native `FMicrosoft Equation 3.0DNQE Equation.3D m 1 z perc h 1FMicrosoft Equation 3.0DNQE _1075140420F;l;lOle CompObjRObjInfo  Equation.3A  h A12 +h B12FMicrosoft Equation 3.0DNQE Equation.3Equation Native ]_1075141192F;l;lOle CompObjRObjInfoEquation Native 1_1075141222F;l;lOle   0.14FMicrosoft Equation 3.0DNQE Equation.3 5.15FMicrosoft Equation 3.0DNQE CompObj RObjInfo Equation Native  1_1075125394F;l;lOle CompObjRObjInfoEquation Native 7Equation.3 2x iOh+'0|  , 8 D P\dltOUTLINEUTLtomanEomaomaNormaltoman16aT^js}\I(iQalnHغ[f9m8ucvKNlTfm0| k~l58titT- ]kBӅV(s߬G,+ @ UԩN8 VzVZH`yix P6KsEF& ֞h/t`UC+ ]:=lU NP̭:cdDdB  S A? 2~oHRWΤ$c~-o`!~oHRWΤ$c~@"Pg x]=N@ߌ["EHL""R,EJ) =WJcq 4̮+v}3 |C n`O$7N] [BWHJLJsƚ-GQx}xlVi94_rB iD OPi-T(1 s8FG|SW\g_5[b w~>cuz*\/Ș 6%C{zGufLi@"Pg Ux}U͎64؆m)@&uӕ%Y(^r)^h%VTQnwH}zl?( o8Cߠ1pK B몦9~[ gvv} Xk.EÔW| <_p }8"#pQyT|x08t\3xA/vuw qVK=xڿ鿆g#́gP)R#7'npC8f?vx+o7opJqVptB[t--| ㊚9Hv/@& $s?y͹hdЄՁK ڝ6f<ܾv2 ƼEoh6OԀ{ͫ-hf EXL܃ƛ896p_ .Zڭ|rL[Xr/]q]yNQ4EyB+&V~=4*snjmX5ʪf Jŧ+u84ksށ ]> 6q9n;Q|fQI2aqCB-&}L΀Cg)I?F<@FNe#M<^Y88*1I>.1Pt3fs*d>B0 y&i&=Ne ٍ f&;?X O0 1MR'zHمʊs]?N`* Fԧ:8>uXv^g,(,LgxrarPh岾1Bo {=GHmh'dٳ1iP85qֆ+p32+tls D4uItH=x2- [t-iq*⥄S=W?|U!&9(7rQX;[1=JRH d"NJU2!?d.m:PpiʼJ 1M7.PvXI!$QImvЖŎ<4(.b,A C>Lo`!6tne> Cxcdd``ng```baV fx,F^equ02Ȓcjl@?H 8#ĺ"L@ 0 vo7P b@]._g1L @ZB?T0mWg]CÊ:;0.Qd<%x}SAo0~I0i7$i]ZmG$[9,qZnۉ++\E$m /rl?{`kw䯚"bh,>&{/ ~9zUy=fy.ˌN7a*c>qxqPMԿ=d dmІs/ g)#ϱܤ)5j9"!B*h\mr塶i$l{[踱&xƚxKgZRB1U~W5pǩKgwSΑ4.7YSGCcYa3VQ Vd-R6s9TT:IRg`{r9T$mYy$ʎN̡?ߒwL&x ]B([/vd}6"K.;ӐDU5i2Lu~eR0&5@[\F[ʎ͢*v TEtIqK RX_\Z:Y~w/ ?x}kZ<=Ï7Dd B  S A? 29}%p'}Do`!u9}%p' 'N*9-xڍVOA};8L0D^!9(  !Fbb"?6v6JI1ba(of'{a1o/y r@WZO=2 &FivsEFE9 40+;-ԍvŴOkl<Ϥsƞ6 /WqOڗ2F!._ ~+$bYsmɻ5DaleuJJʚuZ}O.LQuM#xF.f(e6r NYĴյfS,֪LLi,8iXƸVY6#s_,*SYV^NY,rkUecrgʔg(vlXIl|\+[ N}8f_Xtmʺdw<탟~|Q7Ia}hm"%QHDkš}}%dñ6x{GI?A*DC<4,e,vKH 'N*9xVoE~k7u) =$TC` -UH@hN6xky7C(BDB=r? q'nSkw}{of۸P{Vk1^.GM.ܟ G8{{p) r%==8$q@r\pea7 \|.`:/J>q@Xѿ  !%xd9xooCۑ|>!UW1 ˥*8Vo ߷$slen ̗cL k›ּ֞HODln{R_<נ L;&ѧz%YO ]?Q4r {0[oTOY9Uk+GQ=^6/Um&Dևw;7w3 #bRҍq2)A>ɎI2)cDŽ='AOڳO6aUsPS=,q]4)L!ߖ`Wc">N8Sd$H'd$,-;a""K(S L'Y(٣T=,)Q!K_Y[v%a J9E!3#Vs#qq.f 3&ꔵԲ>Gac #sm۶Z(9nxu/a5Fu?'X~ȅn8)~+qr A淟\S(M~_WKEJ4ߎGa0Ie23 $gZJTo8"tauIe9 *OD2 aucR0giT)G)@`fATkjWߏ19h՛n$ϡy9ǝ! IaV0C AFtg81vXxG^i ϝq\qiU"("nIUGOdm 3&aQ$B1UQ)X8C8՚J4dԬ bzD".RmQ !xӵ)B%LKLÖ83 ;JM2.^Ͳ$nm$t\) 5p La%3 3<.yJk|k=дW6kա 5XjwBN}-6DJePnKYsLИIɴd*~; <%aoDd B  S A? 2}3 Vn_QY{o`!Q3 Vn_Q>)x QxڍSJA1d%Ǻ&qqZoPCE}}UB5_o}W-md}qyHߩ Ɔy%yÿ<^ɔwOɎHNd '^N7_ڻ3QzTžEC|$sJ}CQSRidnWaF >)x x}UnFJvf\IA0!E"@QEs \Z@Ҳ]%M>EoM}>@>;\]|`y9\тf֞j b|@=[xli@>T"KTinýd. = 65?}1c+w{}r>R:}HH߅o]ɏa@2aƪ-w6 aSgS⯃~CKbWfA{#zq,%HZ%aSyYbq6E%Orm0]%ﮁE8" RyUA)sѳusXΒ-.\atBiyT #rJ!\γ,3boc-.1g yTMOfqBm3][6k h&>]ӪXL<{.!FzX6eVA+/D7H^4jEMԉ3y|z*鶮YI(euq\g*SI.2/+S+TYSZAUv A^9(oKLb\d 3b·E0**RWv=jjI9LY*B( \kzA\V1+Gg:W2Hcaz.KP#Ξ m د`dyihҬȡyX-R-fk p]18i \*j"FשIQpM5nX^:Q=gܶӧpoD3>rY!KSMX=znl[t–j5`[wjt{{E5vn^W\DdB  S A? 2;װUrJ>-cފqo`!װUrJ>-cފqX xcdd``.ad``baV fx$",F^equ02Ȓcjl@?H @fzM`d0B2sS;p {~-Č`i iEVF\~Fi_1`1m7W v$S D梻U MaQ!2M4S(J[jb!V, E" FĺXf]IT|]+JlȂ+-AT3 vC?CowS0fߒX x}Tn@iiCQ"h)Q)_HJz :j-&M*U8'G xN<*酑֞o܇#X/\µʑ >}WZ |`AiqGk?d/ (i2AQǰ aL9, D@@sOmXye|6\ڢeKZz}'H-U(jƻjeFR"q!(hܐb33{~LhR{r[\6ϰvBRb"UM1FRH;v_ml-pɳ"ԅ@]zR%q"T]Xb!tʅ)2^+>#~&cG!&C{zGufLi@"Pg Ux}U͎64؆m)@&uӕ%Y(^r)^h%VTQnwH}zl?( o8Cߠ1pK B몦9~[ gvv} Xk.EÔW| <_p }8"#pQyT|x08t\3xA/vuw qVK=xڿ鿆g#́g$d@`,&0*(e112BYpL $ lʛad3|-wrps{~-Č`i iEVF Q~i_4F0TG)I S!2M51 N Ma܏XkC'P"6u.ĆRB%]8_Ll 0a.T&A`XL@ E-TT0Ϡ $Q@bh9n:`W~TF`m|x}UnFRq޺.bbP_,DCNr|iu/hj)HdJwh6z;% 3$%%Fv~fgw_ ȼ;;P&o׋*} S0VbT;-DF"Lуc>Gxoe x LAsccFh^5,-a>Y_J1Mdʼnvw0āme!5Y<7N3`()brw@hZT/^b2)$Zue_ EH*?QWv[B |ΧT/<*3 (Gq.Kɐt d ׾Na#(ŵtCyt_i+wgfnx[tVGi)TV8,(:>:ꬎe r)`þ]؄ЧȱJ^m~U? A:v% Oeq$d@`,&0*(e112BYpL $ lʛad3|-wrX=bF r؁"+#C(?Դ/xMu i` R 2`wc3.7BLej4U3b@\seb] mbCXBيPB[w!6[I eoPve~11 REAb 3 iZ&LP lPR<3 D #T F=3yHC~'v1@8^>TF`m|x}UnFRq޺.bbP_,DCNr|iu/hj)HdJwh6z;% 3$%%Fv~fgw_ ȼ;;P&o׋*} S0VbT;-DF"Lуc>Gxoe x LAsccFh^5,-a>Y_J1Mdʼnvw0āme!5Y<7N3`()brw@hZT/^b2)$Zue_ EH*?QWv[B |ΧT/<*3 (Gq.Kɐt d ׾Na#(ŵtCyt_i+wgfnx[tVGi)TV8,(:>:ꬎe r)`þ]؄ЧȱJ^m~U? A:v% Oeq@prNO&rfEjBL WxcPo N rtڄubԉeb8RV0M X6rb M 9i@V5SU]oЌKPeQ/07^+i^HɥTW6ӐBcFMoe,rTŬr]N)Qs!a;:wMـYBM-Qs),%ja*w) 5_Oq5S aꩺr J [5K$1E`DH֋y|L8Zn&Mfyrr#^mr<>DdTB ! S A!?  28;]䷔f쳗Ho`! ;]䷔f쳗%(+Kxcdd``g```daV fxT,F^equ02Ȓcjl@?H @fzM` ! _\7=b iEVFS~a<= cb04g@bu< C`a]r>?vS[%(+Kx}SKo@;T T*W4 UJ֎"!ĥ;rĪky7iz(W΅? aHa֛DB4H`#GuWEY< zY6~KgUť)U. qb1)ЂC`}x@Nk<^* Գ3|Lo"cE[XH'uXIwMH]$0ibpɲYc3ce72ԦYw{\YԤʙμ'+fN4FU\ RmutssbϷ5KU2k0Ao8$HXLjᶚsn)Qsc•XC5@j\[r(uC:tjH!]0NH~4lx2h \f״mݾoG<ൠoC2úϰ5DdTB " S A"? !2v-+R{o`!sv-+RdK@Cx}OQݻýsB/(P)GH(>w)x{.MfgvBB(+DQMcOͩ9VbC(Z"˥5 fd),UWTd.z9OVvC/e8D*>, I ~]oZK46|R RG~֞sp-CvR`whK@CMx}Sn@;-)%K *\Ј JI8!="!r^8k˻I]JH?Wp| 5̮\i흙7ofwvjChB]6`I+iZO)_!ߊ,}V5GؽYsTH͌aJ1:q8dف3؃ݑr589g:rۻ0ߝ '1]D񜏅*{xa& 9J>&?>E߀"f &y Uh)z$YDH~)"2U2U~mtu'CrIy: aq kbNPȈ+5"rlAfOUfڕ4 aeI82N~%r4*I&2(U11+l}xg]ef",ٰ͘N~]*;ah X_1O*(IʄJ_1{ýPq ᆮUEzYZ;uA"B=*蝢RhcbZz9Wt3I2 #Mq 8o_gX*a1TableSummaryInformation(DocumentSummaryInformation8$CompObjjMicrosoft Word 9.0@b@Kg@h'g@@o/l 6՜.+,0 hp  NISTt*C OUTLINE Title  FMicrosoft Word Document MSWo      !"#$%&'()*+,-./0123456789: i8@8 NormalCJ_HaJmH sH tH @@@ Heading 1$@&CJ4OJPJQJaJFF Heading 2$$@&a$CJ4OJPJQJaJ@@ Heading 3$@&6CJOJPJQJ<< Heading 4$1$H$7$8$@&CJ4<A@< Default Paragraph Font>P> Body Text 2CJ,OJPJQJaJ>Q@> Body Text 3CJ4OJPJQJaJR Body Text Indent 2Nh$d%d&d'dNOPQ^hCJ4OJPJQJaJT^@"T Normal (Web)dd[$\$B*OJPJQJ^Jph, @2, Footer  !&)@A& Page NumbervBvBBstm23`z%efg"#$%@   H   2 r s - . C D F H #$2_`CD^_123YZ!:;dfgh|}!%,4:;BGNW]^chmu|}uwb&'  1 2 R !!!!!I"^"""""J#K#######$8$P$$$.%/%%%R&S&k&&2'3'~'''''''(( ((( (!(&('((()(*(+(,(-(.(/(6(7(8(9(:(;(<(=(>(?(E(F(J(O(P(Q(R(S(T(U(\(`(d(h(m(n(o(p(q(r(t(x(|(}(~(((((((((Z)[))5*6*L*X*************+++++ + + + + +++++++++++++%+&+*+/+0+1+2+3+4+5+<+@+D+H+M+N+O+P+Q+R+T+X+\+]+^+_+`+a+b+c+d+g+t+v+y+|+~+++++++++++++++++++++++++++++++++++++++++++,,,,,, , ,,,,,,,,,,/,0,3,6,8,?,E,L,Q,X,^,e,f,k,l,m,n,o,p,q,r,s,t,{,|,},~,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,- ----%-,---2-3-4-5-6-7-8-9-:-;-B-C-D-E-F-G-H-I-J-K-Q-R-X-]-^-_-`-a-b-c-j-o-t-z----------------------2.3.;.=.D.J.Q.V.W.].a.f.k.p.u.v.w.....////01112222233h3334444,5-5F5T5U5m5n55566666{7}777777888b8c88888c9d9: :::;F;G;P;#<W<X<<<=8=9========> >>>>>>> >">'>+>->.>0>5>9>;><>>>C>H>J>K>M>R>W>Y>Z>\>a>g>j>k>l>>>>>>>3?K?L?M?U?,@-@`@a@@@@@@@@@@@@@@@@AAAA&A'A+A1A@AAABAlAmAAAAARBSBTBbBnBoBwB000000000000000000000000000000000000000000000000 0 0 00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000@0)000000000000000000000000000000@0)0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000>000000000000i@0i@0i@0i@000000000000000000000000000000@0@0@0 0! v6 &* 06W9>6ACF#IJPKR368:;?DJLN[ D h 2!x!!$ (022222233#3$343537777<7>7h7j77777+8,898:8I8J8b8c8888888889999D9E9W9X999999999::::::b<BFALLLLLPQ(RKR479<=>@ABCEFGHIKMOPQRSTUVWXYZ\]^_`abcdefghijklmnopqrstuvwxyz{|}~JR5.0`tv !#Macukl   G[]!57 - / : N P $$ $8$L$N$S&g&i&b*v*x****---000000111222222233444U5i5k57%7'7}777777888888:::.;B;D;;<<1<E<G<>>>vB:::::::::::::::::::::::::::::::::::::::::::::: !!8 @ 0(  B S  ?vBgl6,6667799=:?:D:F:::::::;;;;(;-; >>>>>>>>>>+?-?F?H?_?a?@@TBpBqBwBz}CE  FKD G U _ E J u w  $ ) 3 9 rw&+ |MRZ_'3HN8>a2 7 h m J"N"####$$%%k&n&''))L*V*Y*_***,,--3.5...11o1u122}4444F5I5n5x56688K:R:;;; ;;;#<)<<<===7= >>>>>>3?;???@!@TBpBqBwB333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333Toman$Macintosh HD:NIST:bayesclassday2.docToman$Macintosh HD:NIST:bayesclassday2.docToman<Macintosh HD:Temporary Items:AutoRecovery save of bayesclassToman/Macintosh HD:Temporary Items:Word Work File A 4Toman2Macintosh HD:Temporary Items:Word Work File A 2454Toman2Macintosh HD:Temporary Items:Word Work File A 2454Toman2Macintosh HD:Temporary Items:Word Work File A 1046toman2\\WS100\Home\toman\My Documents\bayesclassday2.doctomanL\\ws100\AppData\toman\Microsoft\Word\AutoRecovery save of bayesclassday2.asdtomanL\\ws100\AppData\toman\Microsoft\Word\AutoRecovery save of bayesclassday2.asd7c@ojyYHk ZrYhR#&/h^R&]fO-'9>  ?)U@o)bX-A .\R N/P5 74E sLzC |NT|ZVWkp)XP >ffklPB$m f-}R#NTk A .!%,4:;BGNW]^chmu|}'''''(( ((( (!(&('((()(*(+(,(-(.(/(6(7(8(9(:(;(<(=(>(?(E(F(J(O(P(Q(R(S(T(U(\(`(d(h(m(n(o(p(q(r(t(x(|(}(~((((((X***********+++++ + + + + +++++++++++++%+&+*+/+0+1+2+3+4+5+<+@+D+H+M+N+O+P+Q+R+T+X+\+]+^+_+`+a+b+c+t+v+|+~+++++++++++++++++++++++++++++++++++++++++++,,,,,, , ,,,,,,,,,/,0,6,8,?,E,L,Q,X,^,e,f,k,l,m,n,o,p,q,r,s,t,{,|,},~,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,- ----%-,---2-3-4-5-6-7-8-9-:-;-B-C-D-E-F-G-H-I-J-K-Q-R-X-]-^-_-`-a-b-c-j-o-t-z-----------------2.3.;.=.D.J.Q.V.W.].a.f.k.p.u.v.===> >>>>>>> >">'>+>->.>0>5>9>;><>>>C>H>J>K>M>R>W>Y>Z>\>a>g>j>k>@@@@@@@@@@@@AAAA&A'A+A1A@AAASBTBpBwB@F?I? /F?F?N"N#N$N&N'*+./+?+@vB`` @` `"`H@`*`,`\@`0`d@`6`p@`<`|@`N`@UnknownGz Times New Roman5Symbol3& z Arial3z TimesI& ??Arial Unicode MS"qhtrbrburb 6t!20d*C2QOUTLINEtomantomanrdDocWord.Document.89q