Hi,
I am doing a project with panel data.
I read online that "If the error terms are correlated, then FE is no suitable since inferences may not be correct and you need to model that relationship (probably using random-effects),". However, I see multiple people (including me) running the regression with both FE and robust standard errors, i.e., in Stata with "xtreg" and "fe vce(cluster CompanyNo)". Is this wrong?
I ran the human test and got the following results, hence I should apply fixed effects.
Array
Array
But also I get for robust standard errors, i.e., I should apply these:
Array
So can I apply clustered/robust standard errors and fixed effects?
Thank you in advance!
Best,
Freja
0 Response to FE and robust standard erros
Post a Comment