楼主在做mlogit回归,希望进而计算边际效用,但是需要进行IIA(无关选择独立性假设)检验,也就是用hausman检验。楼主的命令是:mlogit jobsearch ymwork ymwo medu mparty mwork sex birth edu party health hunyinestimates store m1
mlogit jobsearch ymwork ymwo medu mparty mwork sex birth edu party health hunyin if jobsearch!="shichang":jobsearch
estimates store m2
hausman m2 m1, alleqs constant
输出的结果是:
Test: Ho: difference in coefficients not systematic
chi2(0) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 0.00
Prob>chi2 = .(这里是个点)
(V_b-V_B is not positive definite)
请问有哪位大神可以帮忙解释下,是不是没有通过原假设,我不能再往下求边际效应了


雷达卡





京公网安备 11010802022788号







