请教arlionn几个问题 :
1、我在作hausman检验时的结果如下:我应该如何处理?
. hausman fe re
---- Coefficients ---- | (b) (B) (b-B) sqrt(diag(V_b-V_B)) | fe re Difference S.E. -------------+---------------------------------------------------------------- lngdp | -.1155129 -.0361875 -.0793254 .0277251 lnedu | .0861166 .0406681 .0454485 .0134316 lnpro | .0089286 .0100826 -.001154 .000384 lnfin | -.0120904 -.0121083 .0000179 . ------------------------------------------------------------------------------ b = consistent under Ho and Ha; obtained from xtreg B = inconsistent under Ha, efficient under Ho; obtained from xtreg
Test: H difference in coefficients not systematic
chi2(4) = (b-B)'[(V_b-V_B)^(-1)](b-B) = -37.30 chi2<0 ==> model fitted on these data fails to meet the asymptotic assumptions of the Hausman test; see suest for a generalized test
2、有些面板数据模型的教材涉及到固定效应模型中还有变截距模型、变系数模型等的区别,并介绍如何检验识别模型。请问在stata中如何实现?
3、同样的数据,用eviews作的结果R^2为0.99以上,而stata的结果R^2在0.5以下,甚至小到0.0x,请问这是为何啊?


雷达卡

京公网安备 11010802022788号







