发帖
雷达卡
probit后求marginal effects，用mfx和margins得到不同结果，命令和结果如下，该如何选择
mfx, predict(p)
margins, dydx(*)
. ***********probit******
. xi: probit know_op1 i.hi2 sup_hi no_relate_hiop private_com i.pce i.edu age_10 female marital hh_size i.sah chronic01 urban_nbs migrant i.region i.wave, vce(cluster communityID)
Probit regression Number of obs = 7,153
Wald chi2(26) = 909.59
Prob > chi2 = 0.0000
Log pseudolikelihood = -2113.7879 Pseudo R2 = 0.3710
(Std. Err. adjusted for 448 clusters in communityID)
--------------------------------------------------------------------------------
| Robust
know_op1 | Coef. Std. Err. z P>|z| [95% Conf. Interval]
---------------+----------------------------------------------------------------
_Ihi2_1 | .8528435 .1685863 5.06 0.000 .5224203 1.183267
_Ihi2_2 | .5620086 .1903432 2.95 0.003 .1889429 .9350744
_Ihi2_3 | .2111758 .1191595 1.77 0.076 -.0223725 .4447242
_Ihi2_7 | .3166352 .211818 1.49 0.135 -.0985205 .731791
sup_hi | .2386233 .1385296 1.72 0.085 -.0328897 .5101364
no_relate_hiop | -.4402445 .0704835 -6.25 0.000 -.5783896 -.3020995
private_com | -.23211 .0748275 -3.10 0.002 -.3787692 -.0854507
_Ipce_2 | -.0322339 .078004 -0.41 0.679 -.1851189 .1206512
_Ipce_3 | .1439423 .0813542 1.77 0.077 -.015509 .3033937
_Ipce_4 | .223455 .0813939 2.75 0.006 .0639258 .3829842
_Ipce_5 | .3445144 .0893926 3.85 0.000 .1693081 .5197208
_Iedu_2 | -.0351868 .0642408 -0.55 0.584 -.1610965 .0907229
_Iedu_4 | .1795316 .0707434 2.54 0.011 .0408771 .3181861
_Iedu_5 | .2437232 .0709141 3.44 0.001 .104734 .3827123
age_10 | -.0222211 .0297116 -0.75 0.455 -.0804547 .0360125
female | .0402776 .0517201 0.78 0.436 -.061092 .1416472
marital | .1445962 .0689958 2.10 0.036 .0093669 .2798256
hh_size | .0181902 .0162269 1.12 0.262 -.0136139 .0499942
_Isah_3 | -.0323329 .078616 -0.41 0.681 -.1864175 .1217516
_Isah_4 | .1252313 .0806585 1.55 0.121 -.0328564 .2833191
chronic01 | .0719539 .065815 1.09 0.274 -.0570412 .200949
urban_nbs | .2462554 .0855149 2.88 0.004 .0786492 .4138615
migrant | .1429982 .0611365 2.34 0.019 .0231729 .2628235
_Iregion_1 | -.1022717 .0943762 -1.08 0.279 -.2872457 .0827023
_Iregion_2 | -.2173583 .0900003 -2.42 0.016 -.3937558 -.0409609
_Iwave_2013 | 2.307401 .091165 25.31 0.000 2.128721 2.486081
_cons | -.0186498 .2801454 -0.07 0.947 -.5677248 .5304251
--------------------------------------------------------------------------------
.
. ******mfx: marginal effects are same to dprobit but the P value is slightly different**********
. mfx, predict(p)
Marginal effects after probit
y = Pr(know_op1) (predict, p)
= .94889726
------------------------------------------------------------------------------
variable | dy/dx Std. Err. z P>|z| [ 95% C.I. ] X
---------+--------------------------------------------------------------------
_Ihi2_1*| .0547498 .00754 7.26 0.000 .039971 .069528 .126101
_Ihi2_2*| .0394225 .00865 4.56 0.000 .022463 .056382 .059276
_Ihi2_3*| .0242283 .01491 1.62 0.104 -.005002 .053459 .754509
_Ihi2_7*| .0258767 .01301 1.99 0.047 .000384 .05137 .02139
sup_hi*| .0211255 .01029 2.05 0.040 .000962 .041289 .065427
no_re~op*| -.0443826 .0067 -6.62 0.000 -.057516 -.03125 .574724
privat~m*| -.0245692 .00804 -3.06 0.002 -.040318 -.00882 .487068
_Ipce_2*| -.0034365 .00845 -0.41 0.684 -.019992 .013119 .201314
_Ipce_3*| .0140766 .00741 1.90 0.058 -.000451 .028604 .1974
_Ipce_4*| .0211095 .00712 2.97 0.003 .007162 .035057 .207186
_Ipce_5*| .0310333 .00752 4.13 0.000 .016295 .045771 .219628
_Iedu_2*| -.003757 .00699 -0.54 0.591 -.017466 .009952 .199776
_Iedu_4*| .0173414 .00637 2.72 0.006 .00486 .029823 .213477
_Iedu_5*| .0236759 .0067 3.53 0.000 .010538 .036814 .297777
age_10 | -.002332 .00307 -0.76 0.448 -.008358 .003694 6.03004
female*| .0042506 .00551 0.77 0.440 -.006548 .015049 .583531
marital*| .0165034 .00869 1.90 0.058 -.00053 .033536 .855305
hh_size | .0019089 .00172 1.11 0.268 -.001466 .005283 3.60939
_Isah_3*| -.0034069 .00834 -0.41 0.683 -.019762 .012948 .424857
_Isah_4*| .0130544 .00828 1.58 0.115 -.003171 .02928 .461904
chron~01*| .0078098 .00747 1.05 0.296 -.006832 .022452 .785265
urban~bs*| .0247684 .00823 3.01 0.003 .00864 .040897 .384454
migrant*| .0142413 .00593 2.40 0.016 .002617 .025866 .271774
_Iregi~1*| -.0110713 .0104 -1.07 0.287 -.031446 .009303 .317349
_Iregi~2*| -.0243316 .01057 -2.30 0.021 -.04504 -.003623 .324759
_Iw~2013*| .3427805 .01479 23.18 0.000 .313801 .37176 .536838
------------------------------------------------------------------------------
(*) dy/dx is for discrete change of dummy variable from 0 to 1
. ********margins:results are different from mfx********
. margins, dydx(*)
Average marginal effects Number of obs = 7,153
Model VCE : Robust
Expression : Pr(know_op1), predict()
dy/dx w.r.t. : _Ihi2_1 _Ihi2_2 _Ihi2_3 _Ihi2_7 sup_hi no_relate_hiop private_com _Ipce_2 _Ipce_3 _Ipce_4 _Ipce_5
_Iedu_2 _Iedu_4 _Iedu_5 age_10 female marital hh_size _Isah_3 _Isah_4 chronic01 urban_nbs migrant
_Iregion_1 _Iregion_2 _Iwave_2013
--------------------------------------------------------------------------------
| Delta-method
| dy/dx Std. Err. z P>|z| [95% Conf. Interval]
---------------+----------------------------------------------------------------
_Ihi2_1 | .1426389 .0278602 5.12 0.000 .0880338 .1972439
_Ihi2_2 | .0939965 .0316918 2.97 0.003 .0318816 .1561113
_Ihi2_3 | .0353194 .0198865 1.78 0.076 -.0036576 .0742963
_Ihi2_7 | .0529575 .0353638 1.50 0.134 -.0163542 .1222693
sup_hi | .03991 .0231971 1.72 0.085 -.0055556 .0853755
no_relate_hiop | -.0736313 .0112974 -6.52 0.000 -.0957737 -.0514889
private_com | -.0388206 .0123499 -3.14 0.002 -.0630259 -.0146154
_Ipce_2 | -.0053911 .0130446 -0.41 0.679 -.0309582 .0201759
_Ipce_3 | .0240745 .0135809 1.77 0.076 -.0025436 .0506926
_Ipce_4 | .0373731 .0136402 2.74 0.006 .0106387 .0641074
_Ipce_5 | .0576204 .0149916 3.84 0.000 .0282373 .0870034
_Iedu_2 | -.005885 .0107444 -0.55 0.584 -.0269437 .0151737
_Iedu_4 | .0300268 .0117906 2.55 0.011 .0069177 .0531359
_Iedu_5 | .0407629 .0118613 3.44 0.001 .0175151 .0640107
age_10 | -.0037165 .0049508 -0.75 0.453 -.0134198 .0059868
female | .0067365 .008655 0.78 0.436 -.010227 .0237
marital | .0241839 .0115329 2.10 0.036 .0015797 .046788
hh_size | .0030423 .0027168 1.12 0.263 -.0022825 .0083672
_Isah_3 | -.0054077 .0131561 -0.41 0.681 -.0311932 .0203778
_Isah_4 | .0209451 .0134409 1.56 0.119 -.0053987 .0472888
chronic01 | .0120344 .011028 1.09 0.275 -.0095801 .0336488
urban_nbs | .0411864 .014114 2.92 0.004 .0135236 .0688493
migrant | .0239166 .0102094 2.34 0.019 .0039066 .0439266
_Iregion_1 | -.017105 .0156972 -1.09 0.276 -.047871 .0136609
_Iregion_2 | -.0363534 .0148805 -2.44 0.015 -.0655187 -.0071881
_Iwave_2013 | .385915 .0135659 28.45 0.000 .3593263 .4125037
--------------------------------------------------------------------------------
京公网安备 11010802022788号
