[Gretl-users] The issue of endogeneity in Panel fixed effect regression
Hi all,
I have a small qualm. In my understanding, panel fixed effect regressions take
account of endogeneity in assuming that variables are correlated with the error term. Am
I right in this assertion?. More so, where do 2SLS and GMM come in to out-perform panel
regressions by fixed effects?.
Thank you
Attachments:
- attachment.html (text/html — 502 bytes)