请教各位前辈:
下面的Probit regression (类似逻辑回归)结果是否都没有意义?
因为P>z 十分显著么?
如何判断一个变量对被解释变量有影响?
0.000代表了什么?
P-value为0.080可以接受么?
不胜感激!
Probit regression, reporting marginal effects Number of obs = 160
LR chi2(16) = 56.02
Prob > chi2 = 0.0000
Log likelihood = -65.114068 Pseudo R2 = 0.3008
joine~o2 dF/dx Std. Err. z P>z x-bar [ 95% C.I. ]
exptes~1* -.0312603 .124629 -0.26 0.797 .18125 -.275529 .213008
exptes~3* .1935535 .085973 1.91 0.056 .3125 .02505 .362057
exptes~4* .030532 .1060266 0.28 0.778 .275 -.177276 .23834
gender2* .1415474 .0775673 1.75 0.080 .43125 -.010482 .293576
fqtesco5* .3495452 .0683579 4.03 0.000 .2375 .215566 .483524
fqtesco3* .311378 .0624321 3.66 0.000 .18125 .189013 .433743
fqtesco1* .5022928 .1002085 4.50 0.000 .44375 .305888 .698698
atfavo~1* .0402232 .1804458 0.22 0.823 .5625 -.313444 .39389
atfavo~2* -.2036634 .2248995 -0.99 0.324 .18125 -.644458 .237132
atfavo~4* .0074867 .1837127 0.04 0.968 .19375 -.352584 .367557
atfavo~5* -.1174853 .4238093 -0.30 0.761 .0125 -.948136 .713166
ethnic1* -.0073835 .2214004 -0.03 0.973 .0375 -.44132 .426553
ethnic2* -.0513104 .1165314 -0.46 0.647 .16875 -.279708 .177087
ethnic3* .1144307 .0959911 1.07 0.285 .25625 -.073708 .30257
ethnic4* -.1030896 .1595919 -0.70 0.486 .1 -.415884 .209705
ethnic5* -.5571041 .2658681 -1.75 0.080 .01875 -1.0782 -.036012
obs. P .73125
pred. P .7782614 (at x-bar)
(*) dF/dx is for discrete change of dummy variable from 0 to 1
z and P>z correspond to the test of the underlying coefficient being 0



雷达卡


京公网安备 11010802022788号







