The RATS Software Forum

Dear Sir
I have 9 years data for 5 banks. Thus 45 observations. I want to know what is the minimum number of observations required to estimate fixed effect model so that one is confident about coefficients. Thanks

Tahir

That's probably fine. You obviously don't get consistent estimates of the individual-specific intercepts, but assuming the underlying model is correct (that the slope coefficients are fixed across individuals), you have enough to get reasonable estimates of those.