[STATA] - Panel data regression

#1
Hi guys,

I need help with analysing panel data. I used the Fixed effects model, however one is not able to control for serial correlation or heteroskedasticity at the same time. therefore, i went onto using the feasible least squares method but how do i incorporate fixed effects into the fgls method? and i also read somewhere that one cannot use fgls if n>t, which is true in my case. furthermore, if i cant incorporate fixed effects into fgls then how do i control for autocorrelation and heteroskedasticity. ran the hausman test and random effects has been rejected.

thanks in advance
 
#2
Hello

To include fixed effects in you panel data analysis you can try this method (xtfevd is the command in STATA):
Efficient Estimation of Time-Invariant and Rarely Changing Variables in Finite Sample Panel Analyses with Unit Fixed Effects (with Vera E. Troeger)

Daniela