------------------------------------------------------------------------------- log: C:\Stata8\lab27oct.smcl log type: smcl opened on: 27 Oct 2004, 10:00:36 . use "C:\Documents and Settings\computob\Escritorio\Sleep75.dta", clear . desc Contains data from C:\Documents and Settings\computob\Escritorio\Sleep75.dta obs: 706 vars: 34 26 Jan 2000 13:57 size: 49,420 (95.3% of memory free) ------------------------------------------------------------------------------- storage display value variable name type format label variable label ------------------------------------------------------------------------------- age byte %9.0g in years black byte %9.0g =1 if black case int %9.0g identifier clerical float %9.0g =1 if clerical worker construc float %9.0g =1 if construction worker educ byte %9.0g years of schooling earns74 float %9.0g total earnings, 1974 gdhlth byte %9.0g =1 if in good or excel. health inlf byte %9.0g =1 if in labor force leis1 int %9.0g sleep - totwrk leis2 int %9.0g slpnaps - totwrk leis3 int %9.0g rlxall - totwrk smsa byte %9.0g =1 if live in smsa lhrwage float %9.0g log hourly wage lothinc float %9.0g log othinc, unless othinc < 0 male byte %9.0g =1 if male marr byte %9.0g =1 if married prot byte %9.0g =1 if Protestant rlxall int %9.0g slpnaps + personal activs selfe byte %9.0g =1 if self employed sleep int %9.0g mins sleep at night, per wk slpnaps int %9.0g minutes sleep, inc. naps south byte %9.0g =1 if live in south spsepay float %9.0g spousal wage income spwrk75 byte %9.0g =1 if spouse works totwrk int %9.0g mins worked per week union byte %9.0g =1 if belong to union worknrm int %9.0g mins work main job workscnd int %9.0g mins work second job exper byte %9.0g age - educ - 6 yngkid byte %9.0g =1 if children < 3 present yrsmarr byte %9.0g years married hrwage float %9.0g hourly wage agesq int %9.0g age^2 ------------------------------------------------------------------------------- Sorted by: . summ Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- age | 706 38.81586 11.34264 23 65 black | 706 .0495751 .2172193 0 1 case | 706 353.5 203.9489 1 706 clerical | 706 .1823309 .3354132 0 1 construc | 706 .0300751 .148366 0 1 -------------+-------------------------------------------------------- educ | 706 12.78045 2.784702 1 17 earns74 | 706 9767.705 9323.588 0 42500 gdhlth | 706 .8909348 .3119419 0 1 inlf | 706 .7535411 .4312544 0 1 leis1 | 706 4690.724 908.0496 1745 7417 -------------+-------------------------------------------------------- leis2 | 706 4573.996 907.0841 1677 7297 leis3 | 706 4518.785 903.4107 1677 7282 smsa | 706 .3994334 .4901292 0 1 lhrwage | 532 1.430977 .6310362 -1.049822 3.569814 lothinc | 706 6.228292 4.219718 0 10.65728 -------------+-------------------------------------------------------- male | 706 .5665722 .4958996 0 1 marr | 706 .8215297 .3831796 0 1 prot | 706 .6628895 .4730581 0 1 rlxall | 706 3438.295 520.4321 1380 6110 selfe | 706 .131728 .3384346 0 1 -------------+-------------------------------------------------------- sleep | 706 3266.356 444.4134 755 4695 slpnaps | 706 3383.084 499.0469 1335 6110 south | 706 .184136 .3878698 0 1 spsepay | 706 5144.178 8246.899 0 75000 spwrk75 | 706 .48017 .4999608 0 1 -------------+-------------------------------------------------------- totwrk | 706 2122.921 947.4701 0 6415 union | 706 .2181303 .4132692 0 1 worknrm | 706 2093.252 945.3015 0 6415 workscnd | 706 29.66856 148.8343 0 1337 exper | 706 20.03541 12.37752 0 55 -------------+-------------------------------------------------------- yngkid | 706 .1288952 .3353215 0 1 yrsmarr | 706 11.76912 11.59123 0 43 hrwage | 532 5.082839 3.704385 .3500001 35.50999 agesq | 706 1635.144 950.103 529 4225 **¿Cuántas horas duerme por semana la gente? . disp 3266/60 54.433333 ** Regresiones para las horas de sueño... . reg sleep totwrk Source | SS df MS Number of obs = 706 -------------+------------------------------ F( 1, 704) = 81.09 Model | 14381717.2 1 14381717.2 Prob > F = 0.0000 Residual | 124858119 704 177355.282 R-squared = 0.1033 -------------+------------------------------ Adj R-squared = 0.1020 Total | 139239836 705 197503.313 Root MSE = 421.14 ------------------------------------------------------------------------------ sleep | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- totwrk | -.1507458 .0167403 -9.00 0.000 -.1836126 -.117879 _cons | 3586.377 38.91243 92.17 0.000 3509.979 3662.775 ------------------------------------------------------------------------------ . reg sleep totwrk marr Source | SS df MS Number of obs = 706 -------------+------------------------------ F( 2, 703) = 41.40 Model | 14672490.6 2 7336245.3 Prob > F = 0.0000 Residual | 124567345 703 177193.948 R-squared = 0.1054 -------------+------------------------------ Adj R-squared = 0.1028 Total | 139239836 705 197503.313 Root MSE = 420.94 ------------------------------------------------------------------------------ sleep | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- totwrk | -.1502072 .0167379 -8.97 0.000 -.1830695 -.1173449 marr | 53.01728 41.38706 1.28 0.201 -28.23976 134.2743 _cons | 3541.678 52.25272 67.78 0.000 3439.088 3644.268 ------------------------------------------------------------------------------ * Mientras mas trabajas, menos duermes... Los casados duermen más minutos, pero no de * manera significativa (p-value es mayor a 20%) . reg sleep totwrk hrwage Source | SS df MS Number of obs = 532 -------------+------------------------------ F( 2, 529) = 31.03 Model | 10354266 2 5177132.99 Prob > F = 0.0000 Residual | 88272414.4 529 166866.568 R-squared = 0.1050 -------------+------------------------------ Adj R-squared = 0.1016 Total | 98626680.4 531 185737.628 Root MSE = 408.49 ------------------------------------------------------------------------------ sleep | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- totwrk | -.1497769 .0192392 -7.78 0.000 -.1875715 -.1119822 hrwage | .2792611 4.847762 0.06 0.954 -9.243966 9.802488 _cons | 3581.646 48.17896 74.34 0.000 3487.001 3676.292 ------------------------------------------------------------------------------ ** Las horas de sueño no tienen nada que ver con el salario. . reg sleep age educ lhrwage marr male union yngkid Source | SS df MS Number of obs = 532 -------------+------------------------------ F( 7, 524) = 1.71 Model | 2196901.91 7 313843.13 Prob > F = 0.1053 Residual | 96429778.5 524 184026.295 R-squared = 0.0223 -------------+------------------------------ Adj R-squared = 0.0092 Total | 98626680.4 531 185737.628 Root MSE = 428.98 ------------------------------------------------------------------------------ sleep | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | 2.587262 1.825124 1.42 0.157 -.9981974 6.172722 educ | -6.16922 7.63372 -0.81 0.419 -21.16567 8.827233 lhrwage | -17.85512 35.06672 -0.51 0.611 -86.74374 51.0335 marr | 78.10761 49.35272 1.58 0.114 -18.84589 175.0611 male | -78.98617 42.92318 -1.84 0.066 -163.3088 5.336482 union | 20.72474 45.70328 0.45 0.650 -69.05943 110.5089 yngkid | 78.47151 57.58968 1.36 0.174 -34.66349 191.6065 _cons | 3228.435 139.5074 23.14 0.000 2954.373 3502.498 ------------------------------------------------------------------------------ ** Fuera de decir que los hombres duermen menos cuanto mas trabajan, es muy difícil ** predecir las horas de sueño. . clear . use "C:\Documents and Settings\computob\Escritorio\airfare.dta", clear . desc Contains data from C:\Documents and Settings\computob\Escritorio\airfare.dta obs: 1,149 vars: 12 15 Jan 2002 00:29 size: 93,069 (91.1% of memory free) ------------------------------------------------------------------------------- storage display value variable name type format label variable label ------------------------------------------------------------------------------- origin str21 %21s flight's origin destin str24 %24s flight's destination id int %9.0g route identifier dist int %9.0g distance, in miles passen int %9.0g avg. passengers per day fare int %9.0g avg. one-way fare, $ bmktshr float %9.0g fraction market, biggest carrier ldist float %9.0g log(distance) lfare float %9.0g log(fare) ldistsq float %9.0g ldist^2 concen float %9.0g = bmktshr lpassen float %9.0g log(passen) ------------------------------------------------------------------------------- Sorted by: id . summ Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- origin | 0 destin | 0 id | 1149 575 331.832 1 1149 dist | 1149 989.745 612.0313 95 2724 passen | 1149 670.9121 847.3426 2 8497 -------------+-------------------------------------------------------- fare | 1149 188.0235 76.84232 62 522 bmktshr | 1149 .6015722 .1970212 .1797 1 ldist | 1149 6.696482 .6595331 4.553877 7.909857 lfare | 1149 5.152683 .4170387 4.127134 6.257668 ldistsq | 1149 45.27747 8.729749 20.73779 62.56583 -------------+-------------------------------------------------------- concen | 1149 .6015722 .1970212 .1797 1 lpassen | 1149 6.059873 .9166551 .6931472 9.047468 ** Analizando una variable a detalle: . summ fare, detail avg. one-way fare, $ ------------------------------------------------------------- Percentiles Smallest 1% 68 62 5% 80 63 10% 96 65 Obs 1149 25% 132 65 Sum of Wgt. 1149 50% 178 Mean 188.0235 Largest Std. Dev. 76.84232 75% 233 488 90% 291 518 Variance 5904.742 95% 324 520 Skewness .8323212 99% 405 522 Kurtosis 3.911919 ** ¿De qué dependen las tarifas aereas? . reg fare concen dist Source | SS df MS Number of obs = 1149 -------------+------------------------------ F( 2, 1146) = 351.60 Model | 2577717.57 2 1288858.79 Prob > F = 0.0000 Residual | 4200926.79 1146 3665.73019 R-squared = 0.3803 -------------+------------------------------ Adj R-squared = 0.3792 Total | 6778644.37 1148 5904.74248 Root MSE = 60.545 ------------------------------------------------------------------------------ fare | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- concen | 66.28582 10.61892 6.24 0.000 45.45112 87.12052 dist | .0863456 .0034184 25.26 0.000 .0796386 .0930526 _cons | 62.68768 8.827529 7.10 0.000 45.36775 80.00761 ------------------------------------------------------------------------------ * A mayor concentracion de la industria que atiende esa ruta, mayor precio * A mayor distancia, mayor tarifa, pero puede haber una relacion no lineal entre * distancia y tarifa--probemos una variable cuadratica: . generate dist2 = dist^2 . reg fare concen dist dist2 Source | SS df MS Number of obs = 1149 -------------+------------------------------ F( 3, 1145) = 235.38 Model | 2585803.72 3 861934.572 Prob > F = 0.0000 Residual | 4192840.65 1145 3661.86956 R-squared = 0.3815 -------------+------------------------------ Adj R-squared = 0.3798 Total | 6778644.37 1148 5904.74248 Root MSE = 60.513 ------------------------------------------------------------------------------ fare | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- concen | 63.97489 10.72665 5.96 0.000 42.92879 85.02099 dist | .0694106 .0118975 5.83 0.000 .0460672 .0927539 dist2 | 6.55e-06 4.41e-06 1.49 0.138 -2.10e-06 .0000152 _cons | 71.97324 10.81154 6.66 0.000 50.7606 93.18588 ------------------------------------------------------------------------------ ** La distancia al cuadrado tiene un coef positivo, pero no es significativo . reg fare concen dist passen Source | SS df MS Number of obs = 1149 -------------+------------------------------ F( 3, 1145) = 237.39 Model | 2599404.07 3 866468.022 Prob > F = 0.0000 Residual | 4179240.3 1145 3649.99153 R-squared = 0.3835 -------------+------------------------------ Adj R-squared = 0.3819 Total | 6778644.37 1148 5904.74248 Root MSE = 60.415 ------------------------------------------------------------------------------ fare | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- concen | 61.41739 10.78269 5.70 0.000 40.26134 82.57343 dist | .0850374 .003453 24.63 0.000 .0782625 .0918123 passen | -.0052318 .0021464 -2.44 0.015 -.0094431 -.0010206 _cons | 70.42125 9.362524 7.52 0.000 52.05163 88.79088 ------------------------------------------------------------------------------ ** A mayor numero de pasajeros por día, menor tarifa ** Un modelo log-log para las tarifas aereas . reg lfare ldist ldistsq concen lpassen Source | SS df MS Number of obs = 1149 -------------+------------------------------ F( 4, 1144) = 188.24 Model | 79.2517995 4 19.8129499 Prob > F = 0.0000 Residual | 120.40985 1144 .105253366 R-squared = 0.3969 -------------+------------------------------ Adj R-squared = 0.3948 Total | 199.66165 1148 .173921298 Root MSE = .32443 ------------------------------------------------------------------------------ lfare | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- ldist | -1.037317 .2471012 -4.20 0.000 -1.52214 -.5524949 ldistsq | .1084671 .0187113 5.80 0.000 .0717549 .1451794 concen | .2150194 .0581998 3.69 0.000 .1008291 .3292097 lpassen | -.0805355 .010593 -7.60 0.000 -.1013194 -.0597515 _cons | 7.546627 .8182576 9.22 0.000 5.941173 9.152081 ------------------------------------------------------------------------------ ** El modelo log-log permite la interpretacion de los coeficientes como elasticidades ** Noten que la r2 de este modelo (0.3948) es mayor al anterior (0.3819).. ** ¿Es mejor este modelo? No podemos decirlo porque la variable dependiente tambien ** ha cambiado de fare a log(fare)--y ambas variables tienen diferente distribución: . summ fare lfare Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- fare | 1149 188.0235 76.84232 62 522 lfare | 1149 5.152683 .4170387 4.127134 6.257668 * Noten que log(fare) tiene menor coeficiente de variacion (.41/5.25) * que fare (76.8/188)... * ¿Que pasa con la regresion si cambio la escala de la variable dependiente? * Por ejemplo, la tarifa en miles de dolares seria: . replace fare=fare/1000 fare was int now float (1149 real changes made) . reg fare dist concen Source | SS df MS Number of obs = 1149 -------------+------------------------------ F( 2, 1146) = 351.60 Model | 2.57771755 2 1.28885878 Prob > F = 0.0000 Residual | 4.20092679 1146 .00366573 R-squared = 0.3803 -------------+------------------------------ Adj R-squared = 0.3792 Total | 6.77864434 1148 .005904742 Root MSE = .06055 ------------------------------------------------------------------------------ fare | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- dist | .0000863 3.42e-06 25.26 0.000 .0000796 .0000931 concen | .0662858 .0106189 6.24 0.000 .0454511 .0871205 _cons | .0626877 .0088275 7.10 0.000 .0453677 .0800076 ------------------------------------------------------------------------------ * Son los mismos coeficientes que antes (ver abajo), pero 1000 veces mas pequeños! * Noten que la significancia y la r2 no cambiaron en absoluto. * Volviendo al modelo original con tarifa en dolares: . replace fare=fare*000 (1149 real changes made) . use "C:\Documents and Settings\computob\Escritorio\airfare.dta", clear . reg fare dist concen Source | SS df MS Number of obs = 1149 -------------+------------------------------ F( 2, 1146) = 351.60 Model | 2577717.57 2 1288858.79 Prob > F = 0.0000 Residual | 4200926.79 1146 3665.73019 R-squared = 0.3803 -------------+------------------------------ Adj R-squared = 0.3792 Total | 6778644.37 1148 5904.74248 Root MSE = 60.545 ------------------------------------------------------------------------------ fare | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- dist | .0863456 .0034184 25.26 0.000 .0796386 .0930526 concen | 66.28582 10.61892 6.24 0.000 45.45112 87.12052 _cons | 62.68768 8.827529 7.10 0.000 45.36775 80.00761 ------------------------------------------------------------------------------ . clear . use "C:\Documents and Settings\computob\Escritorio\CRIME1.DTA", clear . desc Contains data from C:\Documents and Settings\computob\Escritorio\CRIME1.DTA obs: 2,725 vars: 16 6 Nov 1996 10:54 size: 122,625 (88.3% of memory free) ------------------------------------------------------------------------------- storage display value variable name type format label variable label ------------------------------------------------------------------------------- narr86 byte %9.0g # times arrested, 1986 nfarr86 byte %9.0g # felony arrests, 1986 nparr86 byte %9.0g # property crme arr., 1986 pcnv float %9.0g proportion of prior convictions avgsen float %9.0g avg sentence length, mos. tottime float %9.0g time in prison since 18 (mos.) ptime86 byte %9.0g mos. in prison during 1986 qemp86 float %9.0g # quarters employed, 1986 inc86 float %9.0g legal income, 1986, $100s durat float %9.0g recent unemp duration black byte %9.0g =1 if black hispan byte %9.0g =1 if Hispanic born60 byte %9.0g =1 if born in 1960 pcnvsq float %9.0g pcnv^2 pt86sq int %9.0g ptime86^2 inc86sq float %9.0g inc86^2 ------------------------------------------------------------------------------- Sorted by: . summ born Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- born60 | 2725 .3625688 .48083 0 1 ** Esta base de datos contiene variables sobre los determinantes del crimen y la ** duracion de las sentencias ** Matriz de correlaciones . corr avgsen narr86 inc86 pcnv black hispan born60 (obs=2725) | avgsen narr86 inc86 pcnv black hispan born60 -------------+--------------------------------------------------------------- avgsen | 1.0000 narr86 | 0.0293 1.0000 inc86 | -0.0958 -0.1900 1.0000 pcnv | 0.0258 -0.0725 -0.0089 1.0000 black | 0.1194 0.1493 -0.1470 -0.0661 1.0000 hispan | 0.0128 0.0530 0.0008 0.0102 -0.2311 1.0000 born60 | 0.0123 -0.0245 0.0811 0.0473 0.0100 -0.0111 1.0000 . * Pairwise correlation matrix - correlaciones de pares con grado de significancia . pwcorr avgsen narr86 inc86 pcnv black hispan born60, sig | avgsen narr86 inc86 pcnv black hispan born60 -------------+--------------------------------------------------------------- avgsen | 1.0000 | | narr86 | 0.0293 1.0000 | 0.1263 | inc86 | -0.0958 -0.1900 1.0000 | 0.0000 0.0000 | pcnv | 0.0258 -0.0725 -0.0089 1.0000 | 0.1776 0.0002 0.6427 | black | 0.1194 0.1493 -0.1470 -0.0661 1.0000 | 0.0000 0.0000 0.0000 0.0006 | hispan | 0.0128 0.0530 0.0008 0.0102 -0.2311 1.0000 | 0.5053 0.0056 0.9679 0.5945 0.0000 | born60 | 0.0123 -0.0245 0.0811 0.0473 0.0100 -0.0111 1.0000 | 0.5219 0.2013 0.0000 0.0136 0.6006 0.5623 | * El numero debajo del coeficiente es la significancia de la correlacion--y debe ser * menora 0.10 o 0.05 para ser significativo. * Pidiendo ayuda sobre este (y cualquier otro comando) . help pwcorr * Matriz de correlaciones con significancia, num de obs, y estrellitas al 10%: . pwcorr avgsen narr86 inc86 pcnv black hispan born60, sig obs star(10) | avgsen narr86 inc86 pcnv black hispan born60 -------------+--------------------------------------------------------------- avgsen | 1.0000 | | 2725 | narr86 | 0.0293 1.0000 | 0.1263 | 2725 2725 | inc86 | -0.0958* -0.1900* 1.0000 | 0.0000 0.0000 | 2725 2725 2725 | pcnv | 0.0258 -0.0725* -0.0089 1.0000 | 0.1776 0.0002 0.6427 | 2725 2725 2725 2725 | black | 0.1194* 0.1493* -0.1470* -0.0661* 1.0000 | 0.0000 0.0000 0.0000 0.0006 | 2725 2725 2725 2725 2725 | hispan | 0.0128 0.0530* 0.0008 0.0102 -0.2311* 1.0000 | 0.5053 0.0056 0.9679 0.5945 0.0000 | 2725 2725 2725 2725 2725 2725 | born60 | 0.0123 -0.0245 0.0811* 0.0473* 0.0100 -0.0111 1.0000 | 0.5219 0.2013 0.0000 0.0136 0.6006 0.5623 | 2725 2725 2725 2725 2725 2725 2725 | * Matriz de correlaciones con significancia, num de obs, y estrellitas al 5%: . pwcorr avgsen narr86 inc86 pcnv black hispan born60, sig obs star(5) | avgsen narr86 inc86 pcnv black hispan born60 -------------+--------------------------------------------------------------- avgsen | 1.0000 | | 2725 | narr86 | 0.0293 1.0000 | 0.1263 | 2725 2725 | inc86 | -0.0958* -0.1900* 1.0000 | 0.0000 0.0000 | 2725 2725 2725 | pcnv | 0.0258 -0.0725* -0.0089 1.0000 | 0.1776 0.0002 0.6427 | 2725 2725 2725 2725 | black | 0.1194* 0.1493* -0.1470* -0.0661* 1.0000 | 0.0000 0.0000 0.0000 0.0006 | 2725 2725 2725 2725 2725 | hispan | 0.0128 0.0530* 0.0008 0.0102 -0.2311* 1.0000 | 0.5053 0.0056 0.9679 0.5945 0.0000 | 2725 2725 2725 2725 2725 2725 | born60 | 0.0123 -0.0245 0.0811* 0.0473* 0.0100 -0.0111 1.0000 | 0.5219 0.2013 0.0000 0.0136 0.6006 0.5623 | 2725 2725 2725 2725 2725 2725 2725 | ** Regresiones para la duracion promedio de una sentencia: . reg avgsen narr86 inc86 black born60 Source | SS df MS Number of obs = 2725 -------------+------------------------------ F( 4, 2720) = 14.47 Model | 698.40584 4 174.60146 Prob > F = 0.0000 Residual | 32823.9121 2720 12.0676148 R-squared = 0.0208 -------------+------------------------------ Adj R-squared = 0.0194 Total | 33522.318 2724 12.3062841 Root MSE = 3.4738 ------------------------------------------------------------------------------ avgsen | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- narr86 | -.0075728 .0795442 -0.10 0.924 -.163546 .1484004 inc86 | -.0043073 .0010285 -4.19 0.000 -.006324 -.0022905 black | 1.025392 .1845037 5.56 0.000 .6636102 1.387173 born60 | .1297632 .1389278 0.93 0.350 -.1426515 .4021779 _cons | .6598738 .1112976 5.93 0.000 .4416374 .8781102 ------------------------------------------------------------------------------ * Solo el ingreso y la raza resultan significativos ** REGRESION con errores estandar ROBUSTOS para controlar ** problemas de Heteroscedasticidad (varianza no constante a lo largo de la muestra) . reg avgsen narr86 inc86 black born60, robust Regression with robust standard errors Number of obs = 2725 F( 4, 2720) = 10.46 Prob > F = 0.0000 R-squared = 0.0208 Root MSE = 3.4738 ------------------------------------------------------------------------------ | Robust avgsen | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- narr86 | -.0075728 .0905989 -0.08 0.933 -.1852224 .1700768 inc86 | -.0043073 .0007801 -5.52 0.000 -.0058369 -.0027776 black | 1.025392 .288632 3.55 0.000 .4594317 1.591352 born60 | .1297632 .1448944 0.90 0.371 -.154351 .4138774 _cons | .6598738 .105495 6.26 0.000 .4530154 .8667322 ------------------------------------------------------------------------------ . log close log: C:\Stata8\lab27oct.smcl log type: smcl closed on: 27 Oct 2004, 11:00:18 -------------------------------------------------------------------------------