ࡱ> %` dTbjbjٕ ; 0,x)/ (x1, we obtain the answer. (ii) As in part (i), we use the chain rule, which is now more complicated:  EMBED Equation.DSMT4  where x1= z1 and x2=  EMBED Equation.DSMT4 . But (E(y|y> 0,x)/ (x1= (1{1( ((x(/()[x(/(+ ((x(/()]}, (E(y|y> 0,x)/(x2= (2{1( ((x(/()[x(/(+ ((x(/()]}, (x1/(z1= 1, and (x2/(z1= 2z1. Plugging these into the first formula and rearranging gives the answer. 17.5 (i) patents is a count variable, and so the Poisson regression model is appropriate. (ii) Because (1 is the coefficient on log(sales), (1 is the elasticity of patents with respect to sales. (More precisely, (1 is the elasticity of E(patents|sales,RD) with respect to sales.) (iii) We use the chain rule to obtain the partial derivative of exp[(0+ (1log(sales)+ (2RD+ (3RD2] with respect to RD:  EMBED Equation.DSMT4  = ((2 + 2(3RD)exp[(0 + (1log(sales)+ (2RD + (3RD2]. A simpler way to interpret this model is to take the log and then differentiate with respect to RD: this gives (2+ 2(3RD, which shows that the semi-elasticity of patents with respect to RD is 100((2+ 2(3RD). 17.7 For the immediate purpose of determining the variables that explain whether accepted applicants choose to enroll, there is not a sample selection problem. The population of interest is applicants accepted by the particular university, and you have a random sample from this population. Therefore, it is perfectly appropriate to specify a model for this group, probably a linear probability model, a probit model, or a logit model, and estimate the model using the data at hand. OLS or maximum likelihood estimation will produce consistent, asymptotically normal estimators. This is a good example of where many data analysts knee-jerk reaction might be to conclude that there is a sample selection problem, which is why it is important to be very precise about the purpose of the analysis, which requires one to clearly state the population of interest. If the university is hoping the applicant pool changes in the near future, then there is a potential sample selection problem: the current students that apply may be systematically different from students that may apply in the future. As the nature of the pool of applicants is unlikely to change dramatically over one year, the sample selection problem can be mitigated, if not entirely eliminated, by updating the analysis after each first-year class has enrolled. SOLUTIONS TO COMPUTER EXERCISES C17.1 (i) If spread is zero, there is no favorite, and the probability that the team we (arbitrarily) label the favorite should have a 50% chance of winning. (ii) The linear probability model estimated by OLS gives  EMBED Equation.DSMT4  = .577 + .0194 spread (.028) (.0023) [.032] [.0019] n = 553, R2 = .111. where the usual standard errors are in (() and the heteroskedasticity-robust standard errors are in [(]. Using the usual standard error, the t statistic for H0: (0= .5 is (.577( .5)/.028= 2.75, which leads to rejecting H0 against a two-sided alternative at the 1% level (critical value EMBED Equation.DSMT4  2.58). Using the robust standard error reduces the significance but nevertheless leads to strong rejection of H0 at the 2% level against a two-sided alternative: t= (.577( .5)/.032 EMBED Equation.DSMT4  2.41 (critical value EMBED Equation.DSMT4  2.33). (iii) As we expect, spread is very statistically significant using either standard error, with a t statistic greater than eight. If spread= 10 the estimated probability that the favored team wins is .577+ .0194(10)= .771. (iv) The probit results are given in the following table: Dependent Variable: favwinIndependent VariableCoefficient (Standard Error)spread.0925 (.0122)constant(.0106 (.1037)Number of Observations 553Log Likelihood Value(263.56Pseudo R-Squared.129 In the probit model P(favwin = 1|spread) = (((0 + (1spread), where ((() denotes the standard normal cdf, if (0= 0 then P(favwin = 1|spread) = (((1spread) and, in particular, P(favwin= 1|spread= 0)= ((0)= .5. This is the analog of testing whether the intercept is .5 in the LPM. From the table, the t statistic for testing H0: (0= 0 is only about .102, so we do not reject H0. (v) When spread= 10 the predicted response probability from the estimated probit model is ([.0106+ .0925(10)]= ((.9144) EMBED Equation.DSMT4  .820. This is somewhat above the estimate for the LPM. (vi) When favhome, fav25, and und25 are added to the probit model, the value of the log-likelihood becomes 262.64. Therefore, the likelihood ratio statistic is 2[(262.64 ((263.56)]= 2(263.56 262.64)= 1.84. The p-value from the  EMBED Equation.DSMT4  distribution is about .61, so favhome, fav25, and und25 are jointly very insignificant. Once spread is controlled for, these other factors have no additional power for predicting the outcome. C17.3 (i) Out of 616 workers, 172, or about 18%, have zero pension benefits. For the 444 workers reporting positive pension benefits, the range is from $7.28 to $2,880.27. Therefore, we have a nontrivial fraction of the sample with pensiont= 0, and the range of positive pension benefits is fairly wide. The Tobit model is well-suited to this kind of dependent variable. (ii) The Tobit results are given in the following table: Dependent Variable: pensionIndependent Variable(1)(2)exper5.20 (6.01)4.39 (5.83)age(4.64 (5.71)(1.65 (5.56)tenure36.02 (4.56)28.78 (4.50)educ93.21 (10.89)106.83 (10.77)depends(35.28 (21.92)41.47 (21.21)married(53.69 (71.73)19.75 (69.50)white144.09 (102.08)159.30 (98.97)male308.15 (69.89)257.25 (68.02)union439.05 (62.49)constant(1,252.43 (219.07)(1,571.51 (218.54) Number of Observations616616Log Likelihood Value(3,672.96(3648.55 EMBED Equation.DSMT4 677.74652.90 In column (1), which does not control for union, being white or male (or, of course, both) increases predicted pension benefits, although only male is statistically significant (t EMBED Equation.DSMT4  4.41). (iii) We use equation (17.22) with exper= tenure= 10, age= 35, educ= 16, depends= 0, married= 0, white= 1, and male= 1 to estimate the expected benefit for a white male with the given characteristics. Using our shorthand, we have  EMBED Equation.DSMT4  = (1,252.5 + 5.20(10) 4.64(35) + 36.02(10) + 93.21(16) + 144.09 + 308.15 = 940.90. Therefore, with  EMBED Equation.DSMT4 = 677.74 we estimate E(pension|x) as ((940.9/677.74)((940.9) + (677.74)(((940.9/677.74)  EMBED Equation.DSMT4  966.40. For a nonwhite female with the same characteristics,  EMBED Equation.DSMT4  = (1,252.5 + 5.20(10) 4.64(35) + 36.02(10) + 93.21(16) = 488.66. Therefore, her predicted pension benefit is ((488.66/677.74)((488.66) + (677.74)(((488.66/677.74)  EMBED Equation.DSMT4  582.10. The difference between the white male and nonwhite female is 966.40 582.10= $384.30. (iv) Column (2) in the previous table gives the results with union added. The coefficient is large, but to see exactly how large, we should use equation (17.22) to estimate E(pension|x) with union= 1 and union= 0, setting the other explanatory variables at interesting values. The t statistic on union is over seven. (v) When peratio is used as the dependent variable in the Tobit model, white and male are individually and jointly insignificant. The p-value for the test of joint significance is about .74. Therefore, neither whites nor males seem to have different tastes for pension benefits as a fraction of earnings. White males have higher pension benefits because they have, on average, higher earnings. C17.5 (i) The Poisson regression results are given in the following table: Dependent Variable: kidsIndependent Variable CoefficientStandard Erroreduc(.048.007age.204.055age2(.0022.0006black.360.061east.088.053northcen.142.048west.080.066farm(.015.058othrural(.057.069town.031.049smcity.074.062y74.093.063y76(.029.068y78(.016.069y80(.020.069y82(.193.067y84(.214.069constant(3.0601.211n = 1,129L = (2,070.23 EMBED Equation.DSMT4  = .944 The coefficient on y82 means that, other factors in the model fixed, a womans fertility was about 19.3% lower in 1982 than in 1972. (ii) Because the coefficient on black is so large, we obtain the estimated proportionate difference as exp(.36) 1 EMBED Equation.DSMT4  .433, so a black woman has 43.3% more children than a comparable nonblack woman. (Notice also that black is very statistically significant.) (iii) From the above table,  EMBED Equation.DSMT4 = .944, which shows that there is actually underdispersion in the estimated model. (iv) The sample correlation between kidsi and  EMBED Equation.DSMT4  is about .348, which means the R-squared (or, at least one version of it), is about (.348)2 EMBED Equation.DSMT4  .121. Interestingly, this is actually smaller than the R-squared for the linear model estimated by OLS. (However, remember that OLS obtains the highest possible R-squared for a linear model, while Poisson regression does not obtain the highest possible R-squared for an exponential regression model.) C17.7 (i) When log(wage) is regressed on educ, exper, exper2, nwifeinc, age, kidslt6, and kidsge6, the coefficient and standard error on educ are .0999 (se= .0151). (ii) The Heckit coefficient on educ is .1187 (se= .0341), where the standard error is just the usual OLS standard error. The estimated return to education is somewhat larger than without the Heckit corrections, but the Heckit standard error is over twice as large. (iii) Regressing  EMBED Equation.DSMT4  on educ, exper, exper2, nwifeinc, age, kidslt6, and kidsge6 (using only the selected sample of 428) produces R2 EMBED Equation.DSMT4  .962, which means that there is substantial multicollinearity among the regressors in the second stage regression. This is what leads to the large standard errors. Without an exclusion restriction in the log(wage) equation,  EMBED Equation.DSMT4  is almost a linear function of the other explanatory variables in the sample. C17.9 (i) 248. (ii) The distribution is not continuous: there are clear focal points, and rounding. For example, many more people report one pound than either two-thirds of a pound or 1 1/3 pounds. This violates the latent variable formulation underlying the Tobit model, where the latent error has a normal distribution. Nevertheless, we should view Tobit in this context as a way to possibly improve functional form. It may work better than the linear model for estimating the expected demand function. (ii) The following table contains the Tobit estimates and, for later comparison, OLS estimates of a linear model: Dependent Variable: ecolbsIndependent VariableTobitOLS (Linear Model)ecoprc(5.82 (.89)(2.90 (.59)regprc5.66 (1.06)3.03 (.71)faminc.0066 (.0040).0028 (.0027)hhsize.130 (.095).054 (.064)constant1.00 (.67)1.63 (.45)Number of Observations660660Log Likelihood Value(1,266.44((( EMBED Equation.DSMT4 3.442.48R-squared .0369.0393 Only the price variables, ecoprc and regprc, are statistically significant at the 1% level. (iv) The signs of the price coefficients accord with basic demand theory: the own-price effect is negative, the cross price effect for the substitute good (regular apples) is positive. (v) The null hypothesis can be stated as H0: (1 + (2 = 0. Define (1 = (1 + (2. Then  EMBED Equation.DSMT4  (.16. To obtain the t statistic, I write (2 = (1 ( (1, plug in, and rearrange. This results in doing Tobit of ecolbs on (ecoprc ( regprc), regprc, faminc, and hhsize. The coefficient on regprc is  EMBED Equation.DSMT4  and, of course we get its standard error: about .59. Therefore, the t statistic is about (.27 and p-value = .78. We do not reject the null. (vi) The smallest fitted value is .798, while the largest is 3.327. (vii) The squared correlation between ecolbsi and  EMBED Equation.DSMT4  is about .0369. This is one possible R-squared measure. (viii) The linear model estimates are given in the table for part (ii). The OLS estimates are smaller than the Tobit estimates because the OLS estimates are estimated partial effects on E(ecolbs|x), whereas the Tobit coefficients must be scaled by the term in equation (17.27). The scaling factor is always between zero and one, and often substantially less than one. The Tobit model does not fit better, at least in terms of estimating E(ecolbs|x): the linear model R-squared is a bit larger (.0393 versus .0369). (ix) This is not a correct statement. We have another case where we have confidence in the ceteris paribus price effects (because the price variables are exogenously set), yet we cannot explain much of the variation in ecolbs. The fact that demand for a fictitious product is hard to explain is not very surprising. C17.11 (i) The fraction of women in the work force is 3,286/5,634 ( .583. (ii) The OLS results using the selected sample are  EMBED Equation.DSMT4  = .649 + .099 educ + .020 exper ( .00035 exper2 (.060) (.004) (.003) (.00008) ( .030 black + .014 hispanic (.034) (.036) n = 3,286, R2 = .205 While the point estimates imply blacks earn, on average, about 3% less and Hispanics about 1.3% more than the base group (non-black, non-Hispanic), neither coefficient is statistically significant or even very close to statistical significance at the usual levels. The joint F test gives a p-value of about .63. So, there is little evidence for differences by race and ethnicity once education and experience have been controlled for. (iii) The coefficient on nwifeinc is (.0091 with t = (13.47 and the coefficient on kidlt6 is (.500 with t = (11.05. We expect both coefficients to be negative. If a womans spouse earns more, she is less likely to work. Having a young child in the family also reduces the probability that the woman works. Each variable is very statistically significant. (Not surprisingly, the joint test also yields a p-value of essentially zero.) (iv) We need at least one variable to affect labor force participation that does not have a direct effect on the wage offer. So, we must assume that, controlling for education, experience, and the race/ethnicity variables, other income and the presence of a young children do not affect wage. These propositions could be false if, say, employers discriminate against women who have young children or whose husbands work. Further, if having a young child reduces productivity through, say, having to take time off for sick children and appointments then it would be inappropriate to exclude kidlt6 from the wage equation. (v) The t statistic on the inverse Mills ratio is 1.77 and the p-value against the two-sided alternative is .077. With 3,286 observations, this is not a very small p-value. The test on  EMBED Equation.DSMT4  does not provide strong evidence against the null hypothesis of no selection bias. (vi) Just as important, the slope coefficients do not change much when the inverse Mills ratio is added. For example, the coefficient on educ increases from .099 to .103 a change within the 95% confidence interval for the original OLS estimate. [The 95% CI is (.092,.106.)]. The changes on the experience coefficients are also pretty small; the Heckman estimates are well within the 95% confidence intervals of the OLS estimates. Superficially, the black and hispanic coefficients change by larger amounts, but these estimates are statistically insignificant. Based on the wide confidence intervals, we expect rather wide changes in the estimates to even minor changes in the specification. The most substantial change is in the intercept estimate from .649 to .539 but it is hard to know what to make of this. Remember, in this example, the intercept is the estimated value of log(wage) for a non-black, non-Hispanic woman with zero years of education and experience. No one in the full sample even comes close to this description. Because the slope coefficients do change somewhat, we cannot say that the Heckman estimates imply a lower average wage offer than the uncorrected estimates. Even if this were true, the estimated marginal effects of the explanatory variables are hardly affected. C17.13 (i) Using the entire sample, the estimated coefficient on educ is .1037 with standard error = .0097. (ii) 166 observations are lost when we restrict attention to the sample with educ < 16. This is about 13.5% of the original sample. The coefficient on educ becomes .1182 with standard error = .0126. This is a slight increase in the estimated return to education, and it is estimated less precisely (because we have reduced the sample variation in educ). (iii) If we restrict attention to those with wage < 20, we lose 164 observations [about the same number in part (ii)]. But now the coefficient on educ is much smaller, .0579, with standard error = .0093. (iv) If we use the sample in part (iii) but account for the known truncation point, log(20), the coefficient on educ is .1060 (standard error = .0168). This is very close to the estimate on the original sample. We obtain a less precise estimate because we have dropped 13.3% of the original sample.     PAGE  PAGE 109   !#'012pqr        Q R S T U W X Y [ \ a b c d e g h i j k l s t u v jh#JEHUj,@ h#JCJUVaJjh#JU j-h#Jh#J6H*] h#JH* h#J6] h#J5\h#J h#JCJ B  "#  u v 7 8 23$%    : k & C`$a$?NO{|ϣ j-h#J jbh#J6] h#JH* jh#Jh#J6H*]jC5hRGUhRGUEHUjjF hRGUCJUVaJjh#JU h#J6] h5\ h#J5\hvhf]hnhuNhJ8h#Jh~-4l#$0nfkd?$$IfTl48064 laf4T $$$If v: "$$ v: " IJ}~IJms"W^mvw     "#$% jh#J jbh#J6] jFh#Jh|`j2=h#JEHUj:h#JEHU j-h#J h#J6] h#JH*j28h#JEHUj,@ h#JCJUVaJh#Jjh#JU;09EVW^dlrdT$$ v6$If$$ v6$IfwkdN@$$IfTl40 8j : 064 laf4T $$$$Ifa$ $$$Iflmv}~pp$$ v6$If $$$Ifwkd@$$IfTl40 8j : 064 laf4T~p$$ v6$If $$$IfwkdA$$IfTl40 8j : 064 laf4T|n$$ v6$If $$$IfykdTB$$IfTl40 8j : 064 laf4Tq$$ v6$If$IfykdC$$IfTl40 8j : 064 laf4TWX}~d e 6!~~q~~    : k & C`$a$ykdC$$IfTl40 8j : 064 laf4T %KLMZ`ekqrstu{  - . 0 1 2 a b o u B!I!K!P!V![!!!!!""$"%"<"=">"?"^"e"g"l"ȾjFh#JEHUjW@ h#JCJUVaJ j-h#JjjDh#JEHUj,@ h#JCJUVaJjh#JU jFh#J h#J6] h#JH* jbh#J6]h#J>6!7!##y$z$$$$$$$$$ss$$$ v$Ifa$ $$$IfdkdI$$IfTl4064 laf4T $$ v$If l"r"w"""#######$$$$%%%!%"%/%6%Q%V%t%|%%%%%%% &&&&/&0&B&C&&&&&&&&&&&&&\'a''''''''''jyWh#JEHUj,@ h#JCJUVaJjTh#JEHUjݶ@ h#JCJUVaJjh#JU j-h#Jh#J6H*] h#J5\ hRGU5\ h#J6]h#J>$$$$%%%wn``RB$$ v$If$$ v$If$$ v$If $$$IfkdNJ$$IfTl4F : z06    4 laf4T%%%%!%'%.%zqccUE$$ v$If$$ v$If$$ v$If $$$IfkdK$$IfTlF : z06    4 laT.%/%6%<%C%I%P%zqccUE$$ v$If$$ v$If$$ v$If $$$IfkdK$$IfTlF : z06    4 laTP%Q%V%\%d%k%s%zqccUE$$ v$If$$ v$If$$ v$If $$$IfkdL$$IfTlF : z06    4 laTs%t%|%%%%%zqccUE$$ v$If$$ v$If$$ v$If $$$IfkdHM$$IfTlF : z06    4 laT%%%%%%%zqccUE$$ v$If$$ v$If$$ v$If $$$IfkdN$$IfTlF : z06    4 laT%%%%%%%zqccUE$$ v$If$$ v$If$$ v$If $$$IfkdN$$IfTlF : z06    4 laT%%%%%& &zqccUE$$ v$If$$ v$If$$ v$If $$$IfkdO$$IfTlF : z06    4 laT & &&&&%&zqaSC$$ v$If$$ v$If$$$ v$Ifa$ $$$Ifkd4L4P4b4f4h4m4o4t4u4w4444󊃊 h#J5\ hSe5\jʁh#JEHUj,@ h#JCJUVaJ h#JH*j~h]jh]jEHUjkF h]jCJUVaJh#J6H*]j'|h#JEHUjݶ@ h#JCJUVaJ h#J6]h#Jjh#JU34444444555666666"6$6)6+6061636;6=6@6B6I6O6V6666666666y7}777777777777::::::::ެ j-h#J h#J5\ hC+5\jeh#JEHUjh#JEHUj,@ h#JCJUVaJ h#JH*jJh#JEHUjG@ h#JCJUVaJjh#JU h#J6]h#J;5778899h:i:j:::::~~ $$$Ifikd$$IfTl4N064 laf4T $$ v$If ::::::::bYKK$$ v$If $$$Ifkd$$IfTl4F : I06    4 laf4T$$$ v$Ifa$:::::::::^UGG$$ v$If $$$Ifkdd$$IfTlF : I06    4 laT$$  v : k & C`$If::::; ;9;B;;;;;;;;;;;;;;<< <%=&=(=)=*=-=.=/===>=?=B=C=D=G=H=I=Q=R=i=j=k=l=m=n===========jh#JEHUj@ h#JCJUVaJ jqh#J6] jbh#J6] h#JH*jh#JEHUjݶ@ h#JCJUVaJjh#JU jh#J j-h#Jh#J h#J6]::::; ;;ulUG0$$  v : k & C`$If$$ v$If$$  v : k & C`$If $$$Ifkd!$$IfTlF : I06    4 laT;;; ;%;,;\SEE$$ v$If $$$Ifkdގ$$IfTlF : I06    4 laT$$  v : k & C`$If,;1;8;9;B;G;E<.$$ v$If $$$Ifkd$$IfTlF : I06    4 laT$$  v : k & C`$If$$  v : k & C`$IfG;M;R;X;Y;p;7. $$$IfkdX$$IfTlF :  I 06    4 laT$$  v : k & C`x$If$$  v : k & C`$If$$ v$Ifp;t;x;y;;;;cZL;$$$ v$Ifa$$$ v$If $$$Ifkd#$$IfTlF :  I 06    4 laT$$$ v$Ifa$;;;;;sm__$$ v$If$Ifkd$$IfTlF : I06    4 laT;;;;;smTF$$ v$If$$  v : k & C`r$If$IfkdT$$IfTlF : I06    4 laT;;;=<><<<>>'?(???Asqqqqqqqqqqqqkd$$IfTlF : I06    4 laT =============>>>>>*>0>4>5>L>M>N>O>>>>>>>O?U?V?W?[?\?s?t?u?v???o@u@v@w@lArAsAtAA h#J5\jVhrhrEHUjrmF hrCJUVaJh#J6H*]jh#JEHUjc@ h#JCJUVaJjh#JU j-h#J h#JPJh#J6PJ] h#J6] h#JH* jbh#J6]h#J4AABBBCCDEEGG#J$JOKPK N N& gd N gd-4D5D=DTEVEcEdEEFFFF(F)F,F-FJFOFPFQFTFUF_F`FcFdFGGGGH H$H*H-HHWIcIIIIIIJ J"J#J)J-J.JdJeJJJJJJJļķгjhx&Uhx& hB86hB8hy hV46hephN _h y.h-hEyhQ hk6hk j-hk h+}6h+} h=d6h=dh3 h3H*9JJJJ KNKOKVKmKsKKKKKKKfLLMM"M+MMM N NN!N+N[NNNNN&OgOPoPpPqPwPPP,Q1QvQ{Q:R>RpRtRRRSS;Tǿÿõõñ˭ h$:6 h$:5h$:hhs h6hhh3 h<,6 h&6h&h<,h>&hP>hyhB8hx&jhx&Ujsh/h/EHUjyF h/CJUVaJ7 NpPqPPPARBRSST?TATBTDTETGTHTQTRTST`TaTbTcTdT&`#$ N& gd$: N& gd&;Te@W_Xu0JO* `(n fe*a4N Hnzsǜ UuCn+^d C1V2Bc%)mмΧ:Gĉ )UPfvjNvo%m!ϙ~YgWy\O:+Y9FbzR`Yzd<`6ی =/ۛAGYr|_m20m3q$Xxi^;чwrcL𳋞3ь +띊`j JF7cMݯ)ltNDd ,J   C A? "2ն:KIX$^6-`!ն:KIX$^6r`!x]Q/QmKUeD"ďHܺ JmZlhҪl+P?Yd]HԼ[d7?F@ YQHt]i< ^\J<>e@W_Xu0JO* `(n fe*a4N Hnzsǜ UuCn+^d C1V2Bc%)mмΧ:Gĉ )UPfvjNvo%m!ϙ~YgWy\O:+Y9FbzR`Yzd<`6ی =/ۛAGYr|_m20m3q$Xxi^;чwrcL𳋞3ь +띊`j JF7cMݯ)ltNDd ,J   C A? "2ն:KIX$^6r-`!ն:KIX$^6r`!x]Q/QmKUeD"ďHܺ JmZlhҪl+P?Yd]HԼ[d7?F@ YQHt]i< ^\J<>e@W_Xu0JO* `(n fe*a4N Hnzsǜ UuCn+^d C1V2Bc%)mмΧ:Gĉ )UPfvjNvo%m!ϙ~YgWy\O:+Y9FbzR`Yzd<`6ی =/ۛAGYr|_m20m3q$Xxi^;чwrcL𳋞3ь +띊`j JF7cMݯ)ltNDd ,J   C A? "2ն:KIX$^6 -`!ն:KIX$^6r`!x]Q/QmKUeD"ďHܺ JmZlhҪl+P?Yd]HԼ[d7?F@ YQHt]i< ^\J<>e@W_Xu0JO* `(n fe*a4N Hnzsǜ UuCn+^d C1V2Bc%)mмΧ:Gĉ )UPfvjNvo%m!ϙ~YgWy\O:+Y9FbzR`Yzd<`6ی =/ۛAGYr|_m20m3q$Xxi^;чwrcL𳋞3ь +띊`j JF7cMݯ)ltNDd ,J  C A? "2ն:KIX$^6 -`!ն:KIX$^6r`!x]Q/QmKUeD"ďHܺ JmZlhҪl+P?Yd]HԼ[d7?F@ YQHt]i< ^\J<>e@W_Xu0JO* `(n fe*a4N Hnzsǜ UuCn+^d C1V2Bc%)mмΧ:Gĉ )UPfvjNvo%m!ϙ~YgWy\O:+Y9FbzR`Yzd<`6ی =/ۛAGYr|_m20m3q$Xxi^;чwrcL𳋞3ь +띊`j JF7cMݯ)ltNDd ,J  C A? "2ն:KIX$^67 -`!ն:KIX$^6r`!x]Q/QmKUeD"ďHܺ JmZlhҪl+P?Yd]HԼ[d7?F@ YQHt]i< ^\J<>e@W_Xu0JO* `(n fe*a4N Hnzsǜ UuCn+^d C1V2Bc%)mмΧ:Gĉ )UPfvjNvo%m!ϙ~YgWy\O:+Y9FbzR`Yzd<`6ی =/ۛAGYr|_m20m3q$Xxi^;чwrcL𳋞3ь +띊`j JF7cMݯ)ltNDd @J  C A?   !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~ <Root Entryn F{3Data @WordDocumentm;ObjectPoolp{{_1076683435F{{Ole CompObjiObjInfo "%&'()*+,-./256789:;<=>?BEFGJMNOPQTWXYZ]`abehijmpqruxyz} FMathType 5.0 Equation MathType EFEquation.DSMT49qw(4DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  yEquation Native _1088177291O F{{Ole CompObj i FMathType 5.0 Equation MathType EFEquation.DSMT49q($DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  p ObjInfo Equation Native  _1088177316 F{{Ole  FMathType 5.0 Equation MathType EFEquation.DSMT49q(,DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  q  0CompObjiObjInfoEquation Native _1088177352"F{{Ole CompObjiObjInfoEquation Native  FMathType 5.0 Equation MathType EFEquation.DSMT49q(,DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  q  1_1090067816EF{{Ole  CompObj!iObjInfo# FMathType 5.0 Equation MathType EFEquation.DSMT49q3zhzDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "E(y|Equation Native $_1090067861F{{Ole 0CompObj 1iy>>0,x)"z 1 == "E(y|y>>0,x)"x 1 " "x 1 "z 1 == "E(y|y>>0,x)"x 1 " 1z 1 , FMathType 5.0 Equation MathType EFEquation.DSMT49q3zh|zDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "E(y|ObjInfo!3Equation Native 4_10881788296$F{{Ole @y>>0,x)"z 1 == "E(y|y>>0,x)"x 1 " "x 1 "z 1 ++ "E(y|y>>0,x)"x 2 " "x 2 "z 1 , FMathType 5.0 Equation MathType EFEquation.DSMT49q(4DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  z 12CompObj#%AiObjInfo&CEquation Native D_1090068200@)F{{Ole HCompObj(*IiObjInfo+KEquation Native Ld FMathType 5.0 Equation MathType EFEquation.DSMT49q3Hzzh|zDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A   "E(patents|sales,RD)"RD FMathType 5.0 Equation MathType EFEquation.DSMT49q. DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/ED/AP_1184000691T^.F{{Ole RCompObj-/SiObjInfo0UEquation Native V _1076683926;3F{{Ole [CompObj24\iG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  !favwin FMathType 5.0 Equation MathType EFEquation.DSMT49qw(<DSMT5WinAllBasicCodePagesObjInfo5^Equation Native __10881821698F{{Ole cTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  H" FMathType 5.0 Equation MathType EFEquation.DSMT49q($DSMT5WinAllBasicCodePagesCompObj79diObjInfo:fEquation Native g_1085726151=F{{Times New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  c 32 FMathType 5.0 Equation MathType EFEquation.DSMT49q($DSMT5WinAllBasicCodePagesOle kCompObj<>liObjInfo?nEquation Native oTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  s  FMathType 5.0 Equation MathType EFEquation.DSMT49q3zzh|zDSMT5WinAllBasicCodePages_1090068939BF{{Ole sCompObjACtiObjInfoDvEquation Native w_1090068968',GF{{Ole {CompObjFH|iTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  xb p< FMathType 5.0 Equation MathType EFEquation.DSMT49q3zzh|zDSMT5WinAllBasicCodePagesObjInfoI~Equation Native _1184000953LF{{Ole Times New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  xb  FMathType 5.0 Equation MathType EFEquation.DSMT49q. DSMT5WinAllBasicCodePagesCompObjKMiObjInfoNEquation Native _10874855111QF{{Times New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  ! kids i  FMathType 5.0 Equation MathType EFEquation.DSMT49q"2ArYE'B-`!ArYE'BR xڕRMK@}m~HR@ ~ x7mRl5h%hO ^TO"$Xg2;og7yK]tPR,Q'$EgC| WܺJ^nc~9>(@Uް"ldNupn7G{xo}3"sBe 4ZnGƱSO߽r~بQqṎ' *)٘ͭ%_v,0xZ3Xt-+CV6<#m,ӗ%u`a Ai&Fw(oz031vu0Rj;u chk0SS L{Dd ,J  C A? "2ն:KIX$^6Y-`!ն:KIX$^6r`!x]Q/QmKUeD"ďHܺ JmZlhҪl+P?Yd]HԼ[d7?F@ YQHt]i< ^\J<>e@W_Xu0JO* `(n fe*a4N Hnzsǜ UuCn+^d C1V2Bc%)mмΧ:Gĉ )UPfvjNvo%m!ϙ~YgWy\O:+Y9FbzR`Yzd<`6ی =/ۛAGYr|_m20m3q$Xxi^;чwrcL𳋞3ь +띊`j JF7cMݯ)ltNDd hJ  C A ? "2&V4a-F <-`!&V4a-F <@|xڝRMKP=sYI ht!W6Rl5hK7ܸ?ޯp~+Fν d2 g2'CSZ'&BD*HcalPĪKÌztL+lcak*<{G.X%hD\bF1礜/Fb=NlUNCQ ׀LһB=nRC3ZУM+z?AzԸG+UBQbӹL~4]Қ(n9pGk9[-q*U]/uD<'?ͺ^:%3mf&M?wlFcQU' y;m[|L=F< >,|vZCa78KrhPcv6gr^M zGMş~Dd ,J  C A? " 2ն:KIX$^6-`!ն:KIX$^6r`!x]Q/QmKUeD"ďHܺ JmZlhҪl+P?Yd]HԼ[d7?F@ YQHt]i< ^\J<>e@W_Xu0JO* `(n fe*a4N Hnzsǜ UuCn+^d C1V2Bc%)mмΧ:Gĉ )UPfvjNvo%m!ϙ~YgWy\O:+Y9FbzR`Yzd<`6ی =/ۛAGYr|_m20m3q$Xxi^;чwrcL𳋞3ь +띊`j JF7cMݯ)ltNDd hJ  C A ? " 2G;!q]nݹ8-`!G;!q]nݹ@R|xڝRMka~$5MnbӀx&Y%%m]l 4REE%,|vE7r۽jHcK6QܹIbwT*Dd @J  C A? " 2ArYE'B-`!ArYE'BR xڕRMK@}m~HR@ ~ x7mRl5h%hO ^TO"$Xg2;og7yK]tPR,Q'$EgC| WܺJ^nc~9>(@Uް"ldNupn7G{xo}3"sBe 4ZnGƱSO߽r~بQqṎ' *)٘ͭ%_v,0xZ3Xt-+CV6<#m,ӗ%u`a Ai&Fw(oz031vu0Rj;u chk0SS L{Dd hJ  C A ? " 2&V4a-F <w-`!&V4a-F <@|xڝRMKP=sYI ht!W6Rl5hK7ܸ?ޯp~+Fν d2 g2'CSZ'&BD*HcalPĪKÌztL+lcak*<{G.X%hD\bF1礜/Fb=NlUNCQ ׀LһB=nRC3ZУM+z?AzԸG+UBQbӹL~4]Қ(n9pGk9[-q*U]/uD<'?ͺ^:%3mf&M?wlFcQU' y;m[|L=F< >,|vZCa78KrhPcv6gr^M zGMş~Dd hJ  C A ? " 2G;!q]nݹ("-`!G;!q]nݹ@R|xڝRMka~$5MnbӀx&Y%%m]l 4REE%,|vE7r۽jHcK6QܹIbwT*Dd @J  C A? "2ArYE'B$-`!ArYE'BR xڕRMK@}m~HR@ ~ x7mRl5h%hO ^TO"$Xg2;og7yK]tPR,Q'$EgC| WܺJ^nc~9>(@Uް"ldNupn7G{xo}3"sBe 4ZnGƱSO߽r~بQqṎ' *)٘ͭ%_v,0xZ3Xt-+CV6<#m,ӗ%u`a Ai&Fw(oz031vu0Rj;u chk0SS L{Dd lJ  C A? "2zgDKj%ucVg'-`!NgDKj%uc&#@7xUKSQsumҍBEANIYNmR0`=DT> ރ`J Z;:qAйw|=8|/dd4p@JIαc$IP#*^yFt–PE'J>!~j!+Ol|A}.{8^ }o-G.5 Q!;*dkBvOFp} {"}ݡEX*̷\K>hej0_yH9Lg۔zٸ6n!`ղO ^6ckB-ݟ./TeʚV,8=' #X. lͰv 9 ͅ葕7Ȋ㳶&CxzU fR 6km~VϮ`h XR6DUftt+TF xot/OI%&RvtIx*9NHab<;Rзӧ{a(Ggʥ8"R,4I/;HްpWdžI`Z*F2.s j̶a7W$4#Dt =4݉ݓi=E):juNxQ yo}O{R PW}B@C;Ջ.S|F+;5H1?̈&B~M+Dd J   C A? "2A",-c_!^i+-`!aA",-c_!^8$@7/xڽUOA~3ҥVZ `)&&HCFb!PU-!@L ̩DxbzP4v;[ D nge`d&10ax>,'YfrDCYy GWlPyZ`G*W[m.,+ D&n2l"q{泝N[CՈUGUS!a AN4~`/\.D?'ţ)=P2IՎR4̓u|ҦW>k3x{>/'Ù*JnΐvrQFt˦'&ٴjM]0fsZ_}ZWR YvhHg,`A: *)F fd37Qq p)DR`F4vI7yg<Ap!w0=1P s#]䂹c0"`%ܟ,Gh 9z6+8^j%*py{f8c ):_0Im*U(#)]M~*2Dd  lJ  C A? "2XЈJq>*Ԛ(4pU2-`!hXЈJq>*Ԛ(4`z6xڕSkA~o6[McPZOmvHLRluf+)X,xIģcC=Ijp/At:x^z $ˈĨ(RЕ9 .AtLNU M'|];U@d5 u*n9g},B?X :w{}7qVۃ'Q~g^H3^Y]~P> PR"Ϊ?To1S跆Na8M8#ڃO>Ƿ&~4k$Bmؖ73sZu;&1='w|;ڲP2; ʽoxކ4ݖ UVɨxʹfj= Gӗ֭n"]mo /hDVt9l؄tĊVh"=u_l0VlӶ"  jt H+ OhWv `"%2{C6φvR'gW٘<\#N+ZVMpJ%uAӺQQ$4Dd b  c $A ? ?3"`?29pSU 0C5E`! pSU 0C`\hxuROQ4޵`b MH\<ڋ 4r/-7X'228t?pS܉^b}s}~O`H8WD #!áBOlF2+YBS`؄N'_HI1yVj*4ǓxIqKٽERհwZtx3oø>ݫQ"E9c!b<rKi]X mΨXA 4G›" sDXD^6}%KLq9>K^Lrphqƨ@%2|-3Y[Ћ[c2{?n6eԾ\6ǭnmuZ95q6Z~]hw5h9qTW5_S,J8M!M-3U=)~S!`䬲e3$w0NF"0B15ND*m~@Y=ȩO>vNGDBE9c!b<rKi]X mΨXA 4G›" sDXD^6}%KLq9>K^Lrphqƨ@%2|-3Y[Ћ[c2{?n6eԾ\6ǭnmuZ95q6Z~]hw5h9qTW5_S,J8M!M-3U=)~S!`䬲e3$w0NF"0B15ND*m~@Y=ȩO>vNGDBE9c!b<rKi]X mΨXA 4G›" sDXD^6}%KLq9>K^Lrphqƨ@%2|-3Y[Ћ[c2{?n6eԾ\6ǭnmuZ95q6Z~]hw5h9qTW5_S,J8M!M-3U=)~S!`䬲e3$w0NF"0B15ND*m~@Y=ȩO>vNGDBE9c!b<rKi]X mΨXA 4G›" sDXD^6}%KLq9>K^Lrphqƨ@%2|-3Y[Ћ[c2{?n6eԾ\6ǭnmuZ95q6Z~]hw5h9qTW5_S,J8M!M-3U=)~S!`䬲e3$w0NF"0B15ND*m~@Y=ȩO>vNGDBJv. e +7 )Y1wy930h@.MExJ+W8P,sOHRH)FbU1?I9Rozn-<_ o#d2&OBnMl'7)Oha$Fo~i7ߞuMըN" 5V˘/+A -uZ}#>_RnsoבHƊqV2 Ji;VkƋ]iS*Zw2}bAgqbgZM2b+"jwv䄅z!SdղyM0QpֿT]2$$If!vh5#v:V l406,5/ 4f4T$$If!vh5: 55z#v: #v#vz:V l4065: 55z/ 4f4T$$If!vh5: 55z#v: #v#vz:V l065: 55z/ 4T$$If!vh5: 55z#v: #v#vz:V l065: 55z/ 4T$$If!vh5: 55z#v: #v#vz:V l065: 55z/ 4T$$If!vh5: 55z#v: #v#vz:V l065: 55z/ 4T$$If!vh5: 55z#v: #v#vz:V l065: 55z/ 4T$$If!vh5: 55z#v: #v#vz:V l065: 55z/ 4T$$If!vh5: 55z#v: #v#vz:V l065: 55z/ 4T$$If!vh5: 55z#v: #v#vz:V l065: 55z/ 4T$$If!vh5: 55z#v: #v#vz:V l065: 55z/ 4T$$If!vh5: 55z#v: #v#vz:V l065: 55z/ /  4T$$If!vh5: 55z#v: #v#vz:V l065: 55z/  / 4T$$If!vh5: 55z#v: #v#vz:V l06,5: 55z/ 4TDd J  C A ? "2\$#^>6iWB[TE`!\$#^>6iWBRHxmRMK@4Vk%_ VA۠*ŦxQ+Id"ՓDP<"$gQA2}30 m  1A8h ' ˲{>&!d*=G"Q]&UYC38v e1F&^;~@56NnG0T댓 "-H,gx\ wp7"n "eҙQė?gt{W?$\+#ۍ#\^TMKhru5o90 3y#,j*k^}"<\ZTChD zL:}*[F?]"@FF@)К'̐OHzBZn ZxRe)?'ص{$$If!vh5: 55z#v: #v#vz:V l06,5: 55z/ 4TDd J  C A ? "20tuWE`!0tu*@@xMQMK@}i~HRkA4PЊS6K!jBh*'=y<H(gT2of'ov0%>E9c!b<rKi]X mΨXA 4G›" sDXD^6}%KLq9>K^Lrphqƨ@%2|-3Y[Ћ[c2{?n6eԾ\6ǭnmuZ95q6Z~]hw5h9qTW5_S,J8M!M-3U=)~S!`䬲e3$w0NF"0B15ND*m~@Y=ȩO>vNGDB>m5ߣwo[]ԩUcfC:qԏkR[Vju+hV~b;l"mf6⯦\o6!c/[H,Sl3,}z[()se;iϦgnJA,K&πi &6Fy\W<{ 6&^LFKf+9~DDd J % C A? "!2\$#^>6iWBn`!\$#^>6iWBRHxmRMK@4Vk%_ VA۠*ŦxQ+Id"ՓDP<"$gQA2}30 m  1A8h ' ˲{>&!d*=G"Q]&UYC38v e1F&^;~@56NnG0T댓 "-H,gx\ wp7"n "eҙQė?gt{W?$\+#ۍ#\^TMKhru5o90 3y#,j*k^}"<\ZTChD zL:}*[F?]"@FF@)К'̐OHzBZn ZxRe)?'ص{Dd J & C A? ""20tup`!0tu*@@xMQMK@}i~HRkA4PЊS6K!jBh*'=y<H(gT2of'ov0%>E9c!b<rKi]X mΨXA 4G›" sDXD^6}%KLq9>K^Lrphqƨ@%2|-3Y[Ћ[c2{?n6eԾ\6ǭnmuZ95q6Z~]hw5h9qTW5_S,J8M!M-3U=)~S!`䬲e3$w0NF"0B15ND*m~@Y=ȩO>vNGDB>m5ߣwo[]ԩUcfC:qԏkR[Vju+hV~b;l"mf6⯦\o6!c/[H,Sl3,}z[()se;iϦgnJA,K&πi &6Fy\W<{ 6&^LFKf+9~DDd J  C A ? "20tudE`!0tu*@@xMQMK@}i~HRkA4PЊS6K!jBh*'=y<H(gT2of'ov0%>E9c!b<rKi]X mΨXA 4G›" sDXD^6}%KLq9>K^Lrphqƨ@%2|-3Y[Ћ[c2{?n6eԾ\6ǭnmuZ95q6Z~]hw5h9qTW5_S,J8M!M-3U=)~S!`䬲e3$w0NF"0B15ND*m~@Y=ȩO>vNGDB6iWB`!\$#^>6iWBRHxmRMK@4Vk%_ VA۠*ŦxQ+Id"ՓDP<"$gQA2}30 m  1A8h ' ˲{>&!d*=G"Q]&UYC38v e1F&^;~@56NnG0T댓 "-H,gx\ wp7"n "eҙQė?gt{W?$\+#ۍ#\^TMKhru5o90 3y#,j*k^}"<\ZTChD zL:}*[F?]"@FF@)К'̐OHzBZn ZxRe)?'ص{$$If!vh5#v:V l4=06,5/ 4f4TDd J . C A? ",20tuŦ`!0tu*@@xMQMK@}i~HRkA4PЊS6K!jBh*'=y<H(gT2of'ov0%>E9c!b<rKi]X mΨXA 4G›" sDXD^6}%KLq9>K^Lrphqƨ@%2|-3Y[Ћ[c2{?n6eԾ\6ǭnmuZ95q6Z~]hw5h9qTW5_S,J8M!M-3U=)~S!`䬲e3$w0NF"0B15ND*m~@Y=ȩO>vNGDB6iWBE`!\$#^>6iWBRHxmRMK@4Vk%_ VA۠*ŦxQ+Id"ՓDP<"$gQA2}30 m  1A8h ' ˲{>&!d*=G"Q]&UYC38v e1F&^;~@56NnG0T댓 "-H,gx\ wp7"n "eҙQė?gt{W?$\+#ۍ#\^TMKhru5o90 3y#,j*k^}"<\ZTChD zL:}*[F?]"@FF@)К'̐OHzBZn ZxRe)?'ص{Dd b < c $A? ?3"`?.2R5hKu.`!&5hKu,@ xڝR1oPmj'-"7UX&HT,HOle CompObjPRiObjInfoSEquation Native n($DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  l  FMathType 5.0 Equation MathType EFEquation.DSMT49q_1090095793YVF{{Ole CompObjUWiObjInfoXEquation Native _1090096227[F{{Ole CompObjZ\i3zzh|zDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/!"!APAE%B_AC_A %!AHA_D_E_E_A  q  1 == FMathType 5.0 Equation MathType EFEquation.DSMT49qObjInfo]Equation Native _1184001394Jc`F{{Ole 3zzh|zDSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/!"!APAE%B_AC_A %!AHA_D_E_E_A  q  1 FMathType 5.0 Equation MathType EFEquation.DSMT49qCompObj_aiObjInfobEquation Native _1184002253heF{{. DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/'_!!/!"!APAE%B_AC_A %!AHA_D_E_E_A  !ecolbs i  FMathType 5.0 Equation MathTyOle CompObjdfiObjInfogEquation Native pe EFEquation.DSMT49q.DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  !log(wage)_1184135671jF{{Ole CompObjikiObjInfol FMathType 5.0 Equation MathType EFEquation.DSMT49q.DSMT5WinAllBasicCodePagesTimes New RomanSymbolCourier NewMT Extra!/ED/APG_APAPAE%B_AC_AE*_HA@AHA*_D_E_E_A  l Oh+'0Equation Native 1TableZSummaryInformation(oDocumentSummaryInformation8     !"#$%&'()*+,-./0123456789:;VAHʒN`d`ҡ? `G`&v"T“wG #ǿ"8Bg3ۜk-0tJG8c[sWuw> ˬ)A5Hؗ8=2%{m><Pb+/[b4ZwzU-;(l; ĚSCk^Вov|ח-S縏(5Ct+( v g00쎒ȕf .ʢ:p4% 'Dd J 0 C A? "/20tu`!0tu*@@xMQMK@}i~HRkA4PЊS6K!jBh*'=y<H(gT2of'ov0%>E9c!b<rKi]X mΨXA 4G›" sDXD^6}%KLq9>K^Lrphqƨ@%2|-3Y[Ћ[c2{?n6eԾ\6ǭnmuZ95q6Z~]hw5h9qTW5_S,J8M!M-3U=)~S!`䬲e3$w0NF"0B15ND*m~@Y=ȩO>vNGDBE9c!b<rKi]X mΨXA 4G›" sDXD^6}%KLq9>K^Lrphqƨ@%2|-3Y[Ћ[c2{?n6eԾ\6ǭnmuZ95q6Z~]hw5h9qTW5_S,J8M!M-3U=)~S!`䬲e3$w0NF"0B15ND*m~@Y=ȩO>vNGDB6iWB`!\$#^>6iWBRHxmRMK@4Vk%_ VA۠*ŦxQ+Id"ՓDP<"$gQA2}30 m  1A8h ' ˲{>&!d*=G"Q]&UYC38v e1F&^;~@56NnG0T댓 "-H,gx\ wp7"n "eҙQė?gt{W?$\+#ۍ#\^TMKhru5o90 3y#,j*k^}"<\ZTChD zL:}*[F?]"@FF@)К'̐OHzBZn ZxRe)?'ص{$$If!vh5: 55I#v: #v#vI:V l06,5: 55I/ 4T$$If!vh5: 55I#v: #v#vI:V l06,5: 55I/ 4TDd J  C A? "2*<>bq8tJyܕf{`!<>bq8tJyܕf{@ChxڍRkQ&Ji݄&T,R/"u%GITИ"J;z{nZ1 䱏޼73}%@BlIX!i[Hc F&=WׂJN;wC|BzŎ${URFvv7m9+ iLhfӗml됫5g.22kg7o/XU?#ە^6Ou4w5ZݬDd J  C A? "2QvD(w'`!QvD(w'RhxڍRM/Q=d>"> JUe;ФU ¢V' ?JI^Gtc2/yGZkȧ B# Dc NUh@ @QQub4߫c<4 fZy5>1dEv'vrJ%,DbV*'|BZ58K\]OU$j2Zu2o+ۉ>5$/8kN=_(G3CX \]ZvT]B:REtGJEͬ&3ir)kCa.`G/io+8y\ Dd $b   c $A? ?3"`?2W &݊dНMpS3E`!+ &݊dНMpS<HD hxڕRoP 4Ie'- VJIDRE$ 1 JL&vC:!&`F, 3ؠ̄D{q{wݏGW(A/E RPt>e.՛!#nΝ> Wu{%lگj`m?͊0Et}JI&1wU}S3+sn?Gou:ArћU֝(Zꕷڻ|imrzWX~DV#`apbmRbSd qgI{gdyn2/%mu2g/n" ֩Aipc-WPn351N\O˭z#@Ώ0^ n[;SS=9  8 D P \hpx CHAPTER 17MargyNormal wooldri1175Microsoft Office Word@@'P?@ SS4@*{Y=՜.+,D՜.+,\ hp  (Dept Chem, Michigan State University sK  CHAPTER 17 TitleH 6> MTWinEqns   FMicrosoft Office Word Document MSWordDocWord.Document.89qN@N Normal vdCJ_HaJmH sH tH Z@Z Heading 1$<@&5CJ KH OJQJ\^JaJ :@: Heading 2$@&5\DA@D Default Paragraph FontVi@V  Table Normal :V 44 la (k(No List LOL equ 1(   : k & C`dLOL equ 2(  p: k & `dPOP equ 3+ & : k & ` i /YLdO"d equ 4@ ;  : k & C` gZ F<<j ZO2Z equ 55 0: k & `  #U hOBh equ 6C > p: k & `vEwN -J$ 4@R4 Header  !4 @b4 Footer  !.)@q. Page NumberLOL MTDisplayEquation vH$4/@4 Listh^h`82@8 List 2^`V>@V Title$<@&a$5CJ KHOJQJ\^JaJ 2B@2 Body TextxHC@H Body Text Indenthx^hRZ@R Plain Text vdCJOJQJ^JaJdL "#uv7823$%2 3 N O   S)*+KL%&Yl#$09EVW^dlmv}WX}~de67yz!'./6<CIPQV\dkst|  %&/9BLUVWnrvw ! !Z![!!!!!K"L"x"y""",#-#o$p$%%J&K&e&f&r&{&|&&&&&&&&&&&&&&&&&&&&&&&&&&&&'' ''''' '!'*'0'5'6';'@'E'F'M'R'W'X'\'a'f'g'k'q'v'w'{''''''''''''''''''''''''((((())E*F*8,9,,,--//0011h2i2j22222222222222222222223 3333 3%3,3138393B3G3M3R3X3Y3p3t3x3y33333333333333=4>44466'7(77799::D;E;y;z;;;;<%<&<=<><==??#B$BOCPC FpHqHHHAJBJKKL?LALBLDLELGLHLQLRLSL`LaLbLeL0000 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 000 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 00 0 0 0 0 0 0 0 0 0 0 0 0 00 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 00 0 0 00 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 00 0 0 0 0 00 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 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 00 0 000@0 @0 @0 @0 @0 0@0H00@0H00@0H00@0H00@0@0@0@0@0@0H00 2&<=<><==HHAJBJKKeLK00 ҙI00I00I00I00J00H00H0 0H0 0H0 0H0 0 &&&) ,  5 T%l"').'14:=A4DJ;TdT+./0123568?APQVjkoxy{|~%l0l6!$%.%P%s%%%% &%&V&v&&&)e......../ /5/E/W/f/v////////5:::;,;G;p;;;;A NdT,479:;<=>@BCDEFGHIJKLMNORSTUWXYZ[\]^_`abcdefghilmnpqrstuvwz}cT-u7OQ+CEQik24c{}24.0 "<TV '?A$<> !1!3!!!!!"""""'(( )%)'))))u****++-.....///333Q5i5k546L6N6[7s7u7{;;;BBBdL:::::::::::::::::::::::::::::::::::::::::::::: ")!!prad9 : ST  &&!')'@,A,//01Y1^12!2222222222233344 455555555566666*606O7V7"8'8o8w888)9.9l9t9::;;;;;; <<>>CC"E*EyHzHHH,I0IvIzI:J>JJJKK4L:LL>L?L?LALBLDLELGLHLbLeLUW $8=38%* " $ !XZ~8A !J!L!|!}!="A""".#9###))G*O*u/y//03<4?4G4L5r5X6^666m8o8j9l9;;;<0<5<??QCZC^D`D"E*EFFIIJJL>L?L?LALBLDLELGLHLbLeL3333333333333333333333333333333333333333333333333 # !!L"K&(/0j2333L5t5{;$<'<<<L>L?L?LALBLDLELGLHLbLeL4L:LL>L?L?LALBLDLELGLHLbLeLQPC+Ey[}-3Q2Zz6On#$a% y.T0V4}5$:P>XF#JMLZANuNRGU(W*\f]S_N _|`=dkdIf]j\\3+lx&l-J8B+}V.v Sj/>&Cs9<yI57<,r3epn6 Se-k~-B8$u&@#$9VW^lmv!./6CPQVdst|  %&/BVWnrvwK&e&f&{&&&&&&&&&&&&&&&&&&&&&&&&&&&'' ''''' '!'*'0'5'6';'@'E'F'M'R'W'X'\'a'f'g'k'q'v'w'{''''''''''''''''''''''''((j2222222222222222 333 3,38393B3M3X3Y3p3t3x3y3333333333333eL@;L;L;L;LdL@UnknownG: Times New Roman5Symbol3& : ArialEV Blackadder ITC]  MS MinchoArial Unicode MS?5 z Courier New"1hg_ ƘKhp Y= Y= !4dsKsK2QHX?(W2 CHAPTER 17Margywooldri1CompObjq