1.面板随机效应模型的异方差如何检验的?我知道固定效应模型的异方差检验 但是我的模型适合随机效应模型的
Modified wald 好像不能做随机的异方差检验
2.还有如果时间T小于N截面 如果用FGLS 可以吗
3.Cross-sectional time-series FGLS regression
Coefficients: generalized least squares
Panels: heteroskedastic
Correlation: common AR(1) coefficient for all panels (0.3792)
Estimated covariances = 59 Number of obs = 236
Estimated autocorrelations = 1 Number of groups = 59
Estimated coefficients = 4 Time periods = 4
Wald chi2(3) = 10.77
Prob > chi2 = 0.0131
------------------------------------------------------------------------------
jx | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
lzhy | -.0993844 .0533564 -1.86 0.063 -.203961 .0051921
gmccer | .0127747 .0056494 2.26 0.024 .0017021 .0238474
dlccer | -.0055142 .2718475 -0.02 0.984 -.5383255 .5272971
_cons | -.1188035 .1167812 -1.02 0.309 -.3476905 .1100835
上面用FGLS 做的 那个R值为何不见了呢?


雷达卡




京公网安备 11010802022788号







