. use "C:\Documents and Settings\Lee\Desktop\keane.dta", replace . bysort status: summarize status educ black exper -------------------------------------------------------------------------------------------------------------------- -> status = 1 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- status | 2473 1 0 1 1 educ | 2473 13.01375 1.860152 7 19 black | 2473 .2689042 .4434797 0 1 exper | 2473 .25556 .7806535 0 8 -------------------------------------------------------------------------------------------------------------------- -> status = 2 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- status | 3233 2 0 2 2 educ | 3233 11.22858 2.074873 7 19 black | 3233 .5459326 .4979627 0 1 exper | 3233 .8874111 1.383806 0 10 -------------------------------------------------------------------------------------------------------------------- -> status = 3 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- status | 6959 3 0 3 3 educ | 6959 12.11683 2.021657 7 19 black | 6959 .3312258 .4706879 0 1 exper | 6959 2.500359 1.998841 0 10 -------------------------------------------------------------------------------------------------------------------- -> status = . Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- status | 0 educ | 58 11.87931 2.248411 8 16 black | 58 1 0 1 1 exper | 58 2.017241 1.896362 0 7 . mlogit status educ black exper expersq, baseoutcome(3) Iteration 0: log likelihood = -12620.959 Iteration 1: log likelihood = -9833.8173 Iteration 2: log likelihood = -9637.1459 Iteration 3: log likelihood = -9614.3442 Iteration 4: log likelihood = -9613.3558 Iteration 5: log likelihood = -9613.3495 Multinomial logistic regression Number of obs = 12665 LR chi2(8) = 6015.22 Prob > chi2 = 0.0000 Log likelihood = -9613.3495 Pseudo R2 = 0.2383 ------------------------------------------------------------------------------ status | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- 1 | educ | .0753913 .0135241 5.57 0.000 .0488847 .101898 black | -.4374037 .0614975 -7.11 0.000 -.5579365 -.3168709 exper | -2.07251 .0548751 -37.77 0.000 -2.180063 -1.964956 expersq | .2031146 .0092658 21.92 0.000 .184954 .2212752 _cons | -.3861733 .1809858 -2.13 0.033 -.7408991 -.0314476 -------------+---------------------------------------------------------------- 2 | educ | -.2814784 .0125901 -22.36 0.000 -.3061545 -.2568023 black | .5697132 .049183 11.58 0.000 .4733163 .6661101 exper | -.9865727 .0362782 -27.19 0.000 -1.057677 -.9154688 expersq | .0792283 .0063641 12.45 0.000 .0667548 .0917017 _cons | 3.413881 .159229 21.44 0.000 3.101798 3.725965 ------------------------------------------------------------------------------ (status==3 is the base outcome) . mfx, predict(p outcome(1)) Marginal effects after mlogit y = Pr(status==1) (predict, p outcome(1)) = .1095742 ------------------------------------------------------------------------------ variable | dy/dx Std. Err. z P>|z| [ 95% C.I. ] X ---------+-------------------------------------------------------------------- educ | .015372 .00128 12.00 0.000 .01286 .017884 12.0652 black*| -.0563015 .00517 -10.89 0.000 -.066434 -.046169 .373865 exper | -.1741132 .00538 -32.34 0.000 -.184665 -.163561 1.6503 expersq | .0175611 .00098 17.88 0.000 .015636 .019486 6.45156 ------------------------------------------------------------------------------ (*) dy/dx is for discrete change of dummy variable from 0 to 1 . mfx, predict(p outcome(2)) Marginal effects after mlogit y = Pr(status==2) (predict, p outcome(2)) = .25990818 ------------------------------------------------------------------------------ variable | dy/dx Std. Err. z P>|z| [ 95% C.I. ] X ---------+-------------------------------------------------------------------- educ | -.0562911 .00224 -25.08 0.000 -.06069 -.051892 12.0652 black*| .125544 .00942 13.32 0.000 .107073 .144015 .373865 exper | -.1307496 .0069 -18.94 0.000 -.144278 -.117221 1.6503 expersq | .0094555 .00123 7.72 0.000 .007054 .011856 6.45156 ------------------------------------------------------------------------------ (*) dy/dx is for discrete change of dummy variable from 0 to 1 . mfx, predict(p outcome(3)) Marginal effects after mlogit y = Pr(status==3) (predict, p outcome(3)) = .63051762 ------------------------------------------------------------------------------ variable | dy/dx Std. Err. z P>|z| [ 95% C.I. ] X ---------+-------------------------------------------------------------------- educ | .0409191 .00254 16.12 0.000 .035945 .045893 12.0652 black*| -.0692425 .01072 -6.46 0.000 -.090257 -.048228 .373865 exper | .3048628 .0078 39.09 0.000 .289577 .320148 1.6503 expersq | -.0270165 .00134 -20.10 0.000 -.029651 -.024382 6.45156 ------------------------------------------------------------------------------ (*) dy/dx is for discrete change of dummy variable from 0 to 1 . predict p1 if e(sample), outcome(1) (option pr assumed; predicted probability) (58 missing values generated) . predict p2 if e(sample), outcome(2) (option pr assumed; predicted probability) (58 missing values generated) . predict p3 if e(sample), outcome(3) (option pr assumed; predicted probability) (58 missing values generated) . summarize p1 p2 p3 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- p1 | 12665 .1952625 .1990548 .0027033 .713258 p2 | 12665 .2552704 .1770144 .0170673 .8110914 p3 | 12665 .549467 .2782568 .1083249 .9717485 . mlogit, rrr Multinomial logistic regression Number of obs = 12665 LR chi2(8) = 6015.22 Prob > chi2 = 0.0000 Log likelihood = -9613.3495 Pseudo R2 = 0.2383 ------------------------------------------------------------------------------ status | RRR Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- 1 | educ | 1.078306 .0145831 5.57 0.000 1.050099 1.107271 black | .6457107 .0397096 -7.11 0.000 .572389 .7284248 exper | .1258695 .0069071 -37.77 0.000 .1130344 .140162 expersq | 1.225213 .0113526 21.92 0.000 1.203163 1.247667 -------------+---------------------------------------------------------------- 2 | educ | .7546672 .0095013 -22.36 0.000 .7362728 .7735211 black | 1.76776 .0869437 11.58 0.000 1.605309 1.94665 exper | .3728524 .0135264 -27.19 0.000 .3472617 .4003289 expersq | 1.082451 .0068889 12.45 0.000 1.069033 1.096038 ------------------------------------------------------------------------------ (status==3 is the base outcome)