0
$\begingroup$

if I have a unbalanced panel survey data of four years, I wondered if I still needed to apply panel weights when run "reg", as pool ols regressions on it? Or should I apply cross-sectional weights for "reg" command? There are repeated time values within panel, but the same personal IDs for each year. There might be repeated values for personal IDs. Thank you.

$\endgroup$

Your Answer

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

Browse other questions tagged or ask your own question.