是的,用LR检验的结果是:
(log-likelihoods of null models cannot be compared)
likelihood-ratio test LR chi2(10) = 63.52
(Assumption: . nested in heterotest) Prob > chi2 = 0.0000
这是不是说明不存在异方差?
但WALD检验的结果是:
Modified Wald test for groupwise heteroskedasticity
in fixed effect regression model
H0: sigma(i)^2 = sigma^2 for all i
chi2 (11) = 189.26
Prob>chi2 = 0.0000
这个是说明存在异方差吧?
为什么两个结果不一样?我该相信哪个结果呢?