-------------------------------------------------------------------------------- log: C:\Documents and Settings\Patricia_Anderson\My Documents\ECON 20\In > Class\nov11.log log type: text opened on: 11 Nov 2002, 11:11:43 . keep if ky==1 ky not found r(111); . use injury . keep if ky==1 (1524 observations deleted) . reg ldurat afchge highearn afhigh variable afchge not found r(111); . reg ldurat afchnge highearn afhigh Source | SS df MS Number of obs = 5626 -------------+------------------------------ F( 3, 5622) = 39.54 Model | 191.071427 3 63.6904757 Prob > F = 0.0000 Residual | 9055.93393 5622 1.6108029 R-squared = 0.0207 -------------+------------------------------ Adj R-squared = 0.0201 Total | 9247.00536 5625 1.64391206 Root MSE = 1.2692 ------------------------------------------------------------------------------ ldurat | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- afchnge | .0076573 .0447173 0.17 0.864 -.0800058 .0953204 highearn | .2564785 .0474464 5.41 0.000 .1634652 .3494918 afhigh | .1906012 .0685089 2.78 0.005 .0562973 .3249051 _cons | 1.125615 .0307368 36.62 0.000 1.065359 1.185871 ------------------------------------------------------------------------------ . **try a DDD . use injury no; data in memory would be lost r(4); . use injury, clear . gen afky=afchnge*ky . gen hiky=highearn*ky . gen afhiky=afhigh*ky . reg ldurat afchnge highearn ky afky hiky afhiky Source | SS df MS Number of obs = 7150 -------------+------------------------------ F( 6, 7143) = 30.14 Model | 302.267994 6 50.377999 Prob > F = 0.0000 Residual | 11938.8421 7143 1.67140447 R-squared = 0.0247 -------------+------------------------------ Adj R-squared = 0.0239 Total | 12241.1101 7149 1.71228285 Root MSE = 1.2928 ------------------------------------------------------------------------------ ldurat | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- afchnge | .1554504 .06651 2.34 0.019 .0250712 .2858297 highearn | .2591595 .0722596 3.59 0.000 .1175093 .4008096 ky | -.2611373 .0585998 -4.46 0.000 -.3760103 -.1462643 afky | -.1477931 .0806129 -1.83 0.067 -.3058183 .0102321 hiky | -.002681 .0869327 -0.03 0.975 -.1730949 .1677329 afhiky | .1906012 .0697857 2.73 0.006 .0538005 .3274019 _cons | 1.386753 .0495342 28.00 0.000 1.289651 1.483854 ------------------------------------------------------------------------------ OOPS – I left out afhigh during class! The correct model is . reg ldurat afchnge highearn ky afky hiky afhigh afhiky Source | SS df MS Number of obs = 7150 -------------+------------------------------ F( 7, 7142) = 26.09 Model | 305.206357 7 43.6009081 Prob > F = 0.0000 Residual | 11935.9037 7142 1.67122707 R-squared = 0.0249 -------------+------------------------------ Adj R-squared = 0.0240 Total | 12241.1101 7149 1.71228285 Root MSE = 1.2928 ------------------------------------------------------------------------------ ldurat | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- afchnge | .0973808 .0796305 1.22 0.221 -.0587186 .2534802 highearn | .1691388 .0991463 1.71 0.088 -.0252172 .3634948 ky | -.2871215 .0617866 -4.65 0.000 -.4082417 -.1660014 afky | -.0897235 .0917369 -0.98 0.328 -.269555 .090108 hiky | .0873398 .1102977 0.79 0.428 -.1288764 .3035559 afhigh | .1919906 .1447922 1.33 0.185 -.0918449 .4758262 afhiky | -.0013895 .1607305 -0.01 0.993 -.3164689 .31369 _cons | 1.412737 .0532672 26.52 0.000 1.308317 1.517156 ------------------------------------------------------------------------------ The reason there is no effect is that in reality, MI had a similar change in Benefits, so we are subtracting off a similar size change . keep if ky==1 (1524 observations deleted) . reg ldurat afchnge highearn afhigh Source | SS df MS Number of obs = 5626 -------------+------------------------------ F( 3, 5622) = 39.54 Model | 191.071427 3 63.6904757 Prob > F = 0.0000 Residual | 9055.93393 5622 1.6108029 R-squared = 0.0207 -------------+------------------------------ Adj R-squared = 0.0201 Total | 9247.00536 5625 1.64391206 Root MSE = 1.2692 ------------------------------------------------------------------------------ ldurat | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- afchnge | .0076573 .0447173 0.17 0.864 -.0800058 .0953204 highearn | .2564785 .0474464 5.41 0.000 .1634652 .3494918 afhigh | .1906012 .0685089 2.78 0.005 .0562973 .3249051 _cons | 1.125615 .0307368 36.62 0.000 1.065359 1.185871 ------------------------------------------------------------------------------ . **add in other control variables . reg ldurat afchnge highearn afhigh male married head-constr Source | SS df MS Number of obs = 5349 -------------+------------------------------ F( 14, 5334) = 16.37 Model | 358.441775 14 25.602984 Prob > F = 0.0000 Residual | 8341.41147 5334 1.56381917 R-squared = 0.0412 -------------+------------------------------ Adj R-squared = 0.0387 Total | 8699.85325 5348 1.62674892 Root MSE = 1.2505 ------------------------------------------------------------------------------ ldurat | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- afchnge | .0106274 .0449167 0.24 0.813 -.0774276 .0986824 highearn | .1757598 .0517462 3.40 0.001 .0743161 .2772034 afhigh | .2308768 .0695248 3.32 0.001 .0945798 .3671738 male | -.0979407 .0445498 -2.20 0.028 -.1852765 -.0106049 married | .1220995 .0391228 3.12 0.002 .0454027 .1987962 head | -.5139003 .1292776 -3.98 0.000 -.7673371 -.2604634 neck | .2699126 .1614899 1.67 0.095 -.0466736 .5864989 upextr | -.178539 .1011794 -1.76 0.078 -.376892 .0198141 trunk | .1264514 .1090163 1.16 0.246 -.0872651 .3401679 lowback | -.0085967 .1015267 -0.08 0.933 -.2076305 .1904371 lowextr | -.1202911 .1023262 -1.18 0.240 -.3208922 .08031 occdis | .2727118 .210769 1.29 0.196 -.1404816 .6859052 manuf | -.1606709 .0409038 -3.93 0.000 -.2408591 -.0804827 construc | .1101967 .0518063 2.13 0.033 .0086352 .2117581 _cons | 1.245922 .1061677 11.74 0.000 1.03779 1.454054 ------------------------------------------------------------------------------ . use crime4 no; data in memory would be lost r(4); . use crime4, clear . desc Contains data from crime4.dta obs: 630 vars: 59 13 Sep 2000 15:29 size: 131,040 (99.9% of memory free) ------------------------------------------------------------------------------- storage display value variable name type format label variable label ------------------------------------------------------------------------------- county int %9.0g county identifier year byte %9.0g 81 to 87 crmrte float %9.0g crimes committed per person prbarr float %9.0g 'probability' of arrest prbconv float %9.0g 'probability' of conviction prbpris float %9.0g 'probability' of prison sentenc avgsen float %9.0g avg. sentence, days polpc float %9.0g police per capita density float %9.0g people per sq. mile taxpc float %9.0g tax revenue per capita west byte %9.0g =1 if in western N.C. central byte %9.0g =1 if in central N.C. urban byte %9.0g =1 if in SMSA pctmin80 float %9.0g perc. minority, 1980 wcon float %9.0g weekly wage, construction wtuc float %9.0g wkly wge, trns, util, commun wtrd float %9.0g wkly wge, whlesle, retail trade wfir float %9.0g wkly wge, fin, ins, real est wser float %9.0g wkly wge, service industry wmfg float %9.0g wkly wge, manufacturing wfed float %9.0g wkly wge, fed employees wsta float %9.0g wkly wge, state employees wloc float %9.0g wkly wge, local gov emps mix float %9.0g offense mix: face-to-face/other pctymle float %9.0g percent young male d82 byte %9.0g =1 if year == 82 d83 byte %9.0g =1 if year == 83 d84 byte %9.0g =1 if year == 84 d85 byte %9.0g =1 if year == 85 d86 byte %9.0g =1 if year == 86 d87 byte %9.0g =1 if year == 87 lcrmrte float %9.0g log(crmrte) lprbarr float %9.0g log(prbarr) lprbconv float %9.0g log(prbconv) lprbpris float %9.0g log(prbpris) lavgsen float %9.0g log(avgsen) lpolpc float %9.0g log(polpc) ldensity float %9.0g log(density) ltaxpc float %9.0g log(taxpc) lwcon float %9.0g log(wcon) lwtuc float %9.0g log(wtuc) lwtrd float %9.0g log(wtrd) lwfir float %9.0g log(wfir) lwser float %9.0g log(wser) lwmfg float %9.0g log(wmfg) lwfed float %9.0g log(wfed) lwsta float %9.0g log(wsta) lwloc float %9.0g log(wloc) lmix float %9.0g log(mix) lpctymle float %9.0g log(pctymle) lpctmin float %9.0g log(pctmin) clcrmrte float %9.0g lcrmrte - lcrmrte[t-1] clprbarr float %9.0g lprbarr - lprbarr[t-1] clprbcon float %9.0g lprbconv - lprbconv[t-1] clprbpri float %9.0g lprbpri - lprbpri[t-1] clavgsen float %9.0g lavgsen - lavgsen[t-1] clpolpc float %9.0g lpolpc - lpolpc[t-1] cltaxpc float %9.0g ltaxpc - ltaxpc[t-1] clmix float %9.0g lmix - lmix[t-1] ------------------------------------------------------------------------------- Sorted by: . reg lcrmrte lprbarr lprbconv lprbpris lavgsen lpolpc d82-d87 Source | SS df MS Number of obs = 630 -------------+------------------------------ F( 11, 618) = 74.49 Model | 117.644666 11 10.6949697 Prob > F = 0.0000 Residual | 88.7356784 618 .14358524 R-squared = 0.5700 -------------+------------------------------ Adj R-squared = 0.5624 Total | 206.380345 629 .328108656 Root MSE = .37893 ------------------------------------------------------------------------------ lcrmrte | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- lprbarr | -.7195033 .0367657 -19.57 0.000 -.7917042 -.6473024 lprbconv | -.5456588 .0263683 -20.69 0.000 -.5974412 -.4938764 lprbpris | .2475521 .0672268 3.68 0.000 .1155314 .3795728 lavgsen | -.0867575 .0579205 -1.50 0.135 -.2005023 .0269873 lpolpc | .3659887 .0300252 12.19 0.000 .3070248 .4249525 d82 | .0051371 .057931 0.09 0.929 -.1086284 .1189026 d83 | -.043503 .0576243 -0.75 0.451 -.1566661 .0696602 d84 | -.1087542 .057923 -1.88 0.061 -.222504 .0049957 d85 | -.0780453 .0583244 -1.34 0.181 -.1925834 .0364928 d86 | -.0420791 .0578218 -0.73 0.467 -.15563 .0714719 d87 | -.0270426 .056899 -0.48 0.635 -.1387815 .0846963 _cons | -2.082293 .2516253 -8.28 0.000 -2.576438 -1.588149 ------------------------------------------------------------------------------ . **correct for hetero . reg lcrmrte lprbarr lprbconv lprbpris lavgsen lpolpc d82-d87, robust Regression with robust standard errors Number of obs = 630 F( 11, 618) = 58.15 Prob > F = 0.0000 R-squared = 0.5700 Root MSE = .37893 ------------------------------------------------------------------------------ | Robust lcrmrte | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- lprbarr | -.7195033 .0517337 -13.91 0.000 -.8210984 -.6179081 lprbconv | -.5456588 .0384277 -14.20 0.000 -.6211236 -.4701941 lprbpris | .2475521 .0852157 2.91 0.004 .0802047 .4148995 lavgsen | -.0867575 .0731524 -1.19 0.236 -.2304148 .0568998 lpolpc | .3659887 .0549735 6.66 0.000 .2580311 .4739462 d82 | .0051371 .0569746 0.09 0.928 -.1067501 .1170243 d83 | -.043503 .0546656 -0.80 0.426 -.1508559 .0638499 d84 | -.1087542 .0572007 -1.90 0.058 -.2210854 .0035771 d85 | -.0780453 .0601414 -1.30 0.195 -.1961517 .0400611 d86 | -.0420791 .0562935 -0.75 0.455 -.1526289 .0684708 d87 | -.0270426 .0581997 -0.46 0.642 -.1413358 .0872506 _cons | -2.082293 .4295719 -4.85 0.000 -2.925891 -1.238696 ------------------------------------------------------------------------------ . reg lcrmrte lprbarr lprbconv lprbpris lavgsen lpolpc d82-d87, cluster(county) Regression with robust standard errors Number of obs = 630 F( 11, 89) = 37.19 Prob > F = 0.0000 R-squared = 0.5700 Number of clusters (county) = 90 Root MSE = .37893 ------------------------------------------------------------------------------ | Robust lcrmrte | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- lprbarr | -.7195033 .1095979 -6.56 0.000 -.9372719 -.5017346 lprbconv | -.5456588 .0704368 -7.75 0.000 -.6856152 -.4057024 lprbpris | .2475521 .1088453 2.27 0.025 .0312787 .4638255 lavgsen | -.0867575 .1130321 -0.77 0.445 -.3113499 .1378349 lpolpc | .3659887 .1210781 3.02 0.003 .1254092 .6065682 d82 | .0051371 .0367296 0.14 0.889 -.0678439 .0781181 d83 | -.043503 .033643 -1.29 0.199 -.1103508 .0233449 d84 | -.1087542 .0391758 -2.78 0.007 -.1865956 -.0309127 d85 | -.0780453 .0385625 -2.02 0.046 -.1546682 -.0014224 d86 | -.0420791 .0428788 -0.98 0.329 -.1272783 .0431201 d87 | -.0270426 .0381447 -0.71 0.480 -.1028354 .0487502 _cons | -2.082293 .8647056 -2.41 0.018 -3.800445 -.3641417 ------------------------------------------------------------------------------ . **simple test for serial correlation . predict uhat, resid . **sort data . sort county year . by county: gen uhat_1=uhat[_n-1] (90 missing values generated) . list county year uhat uhat_1 county year uhat uhat_1 1. 1 81 .1230432 . 2. 1 82 .2164405 .1230432 3. 1 83 .1219607 .2164405 4. 1 84 .4515101 .1219607 5. 1 85 .3703532 .4515101 6. 1 86 .2586627 .3703532 7. 1 87 .2326626 .2586627 8. 3 81 -.1597378 . 9. 3 82 -.283602 -.1597378 10. 3 83 -.2036183 -.283602 11. 3 84 -.2827341 -.2036183 12. 3 85 -.3195593 -.2827341 13. 3 86 -.4766477 -.3195593 14. 3 87 -.3224621 -.4766477 15. 5 81 -1.032502 . 16. 5 82 -.3367048 -1.032502 17. 5 83 -.0613231 -.3367048 18. 5 84 -.0270051 -.0613231 19. 5 85 -.7868816 -.0270051 20. 5 86 -.1529639 -.7868816 21. 5 87 -.7954263 -.1529639 --Break-- r(1); . reg uhat uhat_1 Source | SS df MS Number of obs = 540 -------------+------------------------------ F( 1, 538) = 831.46 Model | 46.6680516 1 46.6680516 Prob > F = 0.0000 Residual | 30.1968236 538 .056127925 R-squared = 0.6071 -------------+------------------------------ Adj R-squared = 0.6064 Total | 76.8648752 539 .142606448 Root MSE = .23691 ------------------------------------------------------------------------------ uhat | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- uhat_1 | .7918086 .02746 28.84 0.000 .7378667 .8457505 _cons | 7.74e-11 .0101951 0.00 1.000 -.0200271 .0200271 ------------------------------------------------------------------------------ . use crime2c no; data in memory would be lost r(4); . use crime2c, clear . reg lcrmrte unem d87 Source | SS df MS Number of obs = 92 -------------+------------------------------ F( 2, 89) = 0.31 Model | .051029791 2 .025514895 Prob > F = 0.7328 Residual | 7.27754362 89 .081770153 R-squared = 0.0070 -------------+------------------------------ Adj R-squared = -0.0154 Total | 7.32857341 91 .080533774 Root MSE = .28595 ------------------------------------------------------------------------------ lcrmrte | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- unem | .0056123 .0113296 0.50 0.622 -.0168994 .0281239 d87 | .0600681 .0760404 0.79 0.432 -.0910225 .2111588 _cons | 4.49762 .121464 37.03 0.000 4.256273 4.738966 ------------------------------------------------------------------------------ . **estimate in changes . clcrmrte cunem unrecognized command: clcrmrte r(199); . reg clcrmrte cunem Source | SS df MS Number of obs = 46 -------------+------------------------------ F( 1, 44) = 5.70 Model | .187053099 1 .187053099 Prob > F = 0.0213 Residual | 1.44309523 44 .032797619 R-squared = 0.1147 -------------+------------------------------ Adj R-squared = 0.0946 Total | 1.63014833 45 .036225519 Root MSE = .1811 ------------------------------------------------------------------------------ clcrmrte | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- cunem | .0189356 .007929 2.39 0.021 .0029558 .0349154 _cons | .1155626 .04247 2.72 0.009 .0299699 .2011553 ------------------------------------------------------------------------------ . **use areg to estimate a fixed effect model . areg lcrmrte unem d87, absorb(code) Number of obs = 92 F( 2, 44) = 3.80 Prob > F = 0.0301 R-squared = 0.9015 Adj R-squared = 0.7964 Root MSE = .12806 ------------------------------------------------------------------------------ lcrmrte | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- unem | .0189356 .007929 2.39 0.021 .0029558 .0349154 d87 | .1155627 .04247 2.72 0.009 .02997 .2011554 _cons | 4.363662 .0819261 53.26 0.000 4.198551 4.528773 -------------+---------------------------------------------------------------- code | F(45, 44) = 8.884 0.000 (46 categories) . xtreg lcrmrte unem d87, i(code) fe Fixed-effects (within) regression Number of obs = 92 Group variable (i) : code Number of groups = 46 R-sq: within = 0.1471 Obs per group: min = 2 between = 0.0008 avg = 2.0 overall = 0.0051 max = 2 F(2,44) = 3.80 corr(u_i, Xb) = -0.1222 Prob > F = 0.0301 ------------------------------------------------------------------------------ lcrmrte | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- unem | .0189356 .007929 2.39 0.021 .0029558 .0349154 d87 | .1155627 .04247 2.72 0.009 .02997 .2011554 _cons | 4.363662 .0819261 53.26 0.000 4.198551 4.528773 -------------+---------------------------------------------------------------- sigma_u | .27221473 sigma_e | .12805783 rho | .81879704 (fraction of variance due to u_i) ------------------------------------------------------------------------------ F test that all u_i=0: F(45, 44) = 8.88 Prob > F = 0.0000 . **what about hetero? . reg clcrmrte cunem, robust Regression with robust standard errors Number of obs = 46 F( 1, 44) = 7.48 Prob > F = 0.0090 R-squared = 0.1147 Root MSE = .1811 ------------------------------------------------------------------------------ | Robust clcrmrte | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- cunem | .0189356 .0069233 2.74 0.009 .0049826 .0328885 _cons | .1155626 .0433426 2.67 0.011 .0282114 .2029138 ------------------------------------------------------------------------------ . areg lcrmrte unem d87, absorb(code) robust Regression with robust standard errors Number of obs = 92 F( 2, 44) = 4.06 Prob > F = 0.0241 R-squared = 0.9015 Adj R-squared = 0.7964 Root MSE = .12806 ------------------------------------------------------------------------------ | Robust lcrmrte | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- unem | .0189356 .0069233 2.74 0.009 .0049827 .0328885 d87 | .1155627 .0433425 2.67 0.011 .0282115 .2029139 _cons | 4.363662 .0748668 58.29 0.000 4.212778 4.514546 -------------+---------------------------------------------------------------- code | absorbed (46 categories) . xtreg lcrmrte unem d87, i(code) fe robust robust invalid r(198); . xtreg lcrmrte unem d87, i(code) re Random-effects GLS regression Number of obs = 92 Group variable (i) : code Number of groups = 46 R-sq: within = 0.1462 Obs per group: min = 2 between = 0.0008 avg = 2.0 overall = 0.0055 max = 2 Random effects u_i ~ Gaussian Wald chi2(2) = 6.50 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0389 ------------------------------------------------------------------------------ lcrmrte | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- unem | .0158128 .0073648 2.15 0.032 .001378 .0302475 d87 | .1025554 .0406813 2.52 0.012 .0228214 .1822894 _cons | 4.39506 .085229 51.57 0.000 4.228015 4.562106 -------------+---------------------------------------------------------------- sigma_u | .25575293 sigma_e | .12805783 rho | .79954609 (fraction of variance due to u_i) ------------------------------------------------------------------------------ . xthaus Hausman specification test ---- Coefficients ---- | Fixed Random lcrmrte | Effects Effects Difference -------------+----------------------------------------- unem | .0189356 .0158128 .0031228 d87 | .1155627 .1025554 .0130073 Test: Ho: difference in coefficients not systematic chi2( 2) = (b-B)'[S^(-1)](b-B), S = (S_fe - S_re) = 1.14 Prob>chi2 = 0.5662 . reg lcrmrte unem d87 Source | SS df MS Number of obs = 92 -------------+------------------------------ F( 2, 89) = 0.31 Model | .051029791 2 .025514895 Prob > F = 0.7328 Residual | 7.27754362 89 .081770153 R-squared = 0.0070 -------------+------------------------------ Adj R-squared = -0.0154 Total | 7.32857341 91 .080533774 Root MSE = .28595 ------------------------------------------------------------------------------ lcrmrte | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- unem | .0056123 .0113296 0.50 0.622 -.0168994 .0281239 d87 | .0600681 .0760404 0.79 0.432 -.0910225 .2111588 _cons | 4.49762 .121464 37.03 0.000 4.256273 4.738966 ------------------------------------------------------------------------------ . log close log: C:\Documents and Settings\Patricia_Anderson\My Documents\ECON 20\I > nClass\nov11.log log type: text closed on: 11 Nov 2002, 12:20:19 -------------------------------------------------------------------------------