Heteroskedasticity test for random effects model

[Thao Nguyen Phuong]

Hi, I have a balanced panel dataset with n = 87 and t = 6
The result of my Hausman test suggests that a random effects model should be used. I would like to ask what is the command to run the test for heteroskedasticity in the random effects model in STATA and how can I fix my model in case it has this problem. My model also has serial autocorrelation and cross-sectional dependence after testing. Thank you so much.

[George Ford]

cluster your errors on the cross section and be done with it. Almost all these models are subject to het and/or AC, and cluster addresses both and causes no harm with n=87.

Here's an option for the test.

https://www.reddit.com/r/stata/comments/5v33ey/hettest_for_panel_data/

https://www.statalist.org/forums/forum/general-stata-discussion/general/1547290-xtreg-re-vs-xtgls

post #11