1
$\begingroup$

I am currently working with a panel data in which I have 6 countries, and the analysis is for 10 years. That is $n=6$ and $t=10$. Is it a good panel?

I am really getting confused as people question me regarding the small number of observations. I can increase the value of $n$, but then the purpose of the paper would really change and there would be huge variation among different countries.

Can someone help me? Please refer to some papers that work with small $n$.

$\endgroup$
3
$\begingroup$

No it is not a good panel. Classical panel data models (random effects, fixed effects) relies on $n$ asymptotics, meaning, that $n$ should be at least 30 (this is rule of thumb, hence not a strict rule). If we are talking about panel VARs, then $t$ asymptotics kick in, and again $T$ should be about 30. What you can do is estimate simple OLS regression without any panel-data flavors, i.e. ignore panel data structure altogether.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.