econfj 發(fā)表于 2013-5-29 15:08
我讀了Maximum likelihood estimation of endogenous switching and sample selection models for binary ...
"paper里面的simulated example: "
經(jīng)測(cè)試這個(gè)是可以運(yùn)行的。
運(yùn)行的結(jié)果,我貼在下面給您看
Sample Selection Probit Regression
(Adaptive quadrature -- 16 points)
Number of obs = 3500
Wald chi2(6) = 1021.27
Log likelihood = -2915.0225 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
y | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
y |
x1 | .3770491 .0509867 7.40 0.000 .2771172 .4769811
x2 | .1400234 .0331205 4.23 0.000 .0751083 .2049384
_cons | .1274961 .0735357 1.73 0.083 -.0166311 .2716234
-------------+----------------------------------------------------------------
selection |
x1 | .9681283 .0342559 28.26 0.000 .900988 1.035269
x2 | .4240503 .027396 15.48 0.000 .370355 .4777455
x3 | -.5845267 .0519791 -11.25 0.000 -.6864038 -.4826495
x4 | .6702432 .0528234 12.69 0.000 .5667113 .7737751
_cons | .4499317 .0430671 10.45 0.000 .3655217 .5343416
-------------+----------------------------------------------------------------
rho | .3929739 .11401 3.45 0.001 .0832102 .5465966
------------------------------------------------------------------------------
Likelihood ratio test for rho=0: chi2(1)= 9.95 Prob>=chi2 = 0.002
"help里面的simulated example:"
請(qǐng)細(xì)看人家建立的資料,
最后一步應(yīng)當(dāng)修正成
ssm
ordvar x1 x2, s(sel = x1 x2 x3 x4) q(15) fam(bin) link(
oprobit) sel
運(yùn)行的結(jié)果,我貼在下面給您看
Sample Selection Ordered Probit Regression
(15 quadrature points)
Number of obs = 3500
Wald chi2(6) = 1165.04
Log likelihood = -5175.5765 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
ordvar | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
ordvar |
x1 | .3154247 .0292513 10.78 0.000 .2580932 .3727563
x2 | .1470224 .0236525 6.22 0.000 .1006643 .1933805
-------------+----------------------------------------------------------------
selection |
x1 | .9573866 .0356374 26.86 0.000 .8875386 1.027235
x2 | .421744 .0286755 14.71 0.000 .365541 .4779469
x3 | -.5968155 .0303954 -19.64 0.000 -.6563894 -.5372415
x4 | .6372247 .0308598 20.65 0.000 .5767405 .6977089
_cons | .5448698 .0288654 18.88 0.000 .4882946 .6014449
-------------+----------------------------------------------------------------
aux_ordvar |
_cut1 | -.4012287 .0460497 -8.71 0.000 -.4914844 -.310973
_cut2 | .1583411 .041944 3.78 0.000 .0761323 .2405498
_cut3 | .4265039 .0409125 10.42 0.000 .3463168 .506691
_cut4 | .787388 .0404784 19.45 0.000 .7080518 .8667242
_cut5 | 1.229028 .0420108 29.26 0.000 1.146688 1.311367
-------------+----------------------------------------------------------------
rho | .3181832 .0688194 4.62 0.000 .1624427 .4320005
------------------------------------------------------------------------------
Likelihood ratio test for rho=0: chi2(1)= 19.42 Prob>=chi2 = 0.000