Panel Regression

[Lucca Mancini]

Hello

At the moment I'm trying to do regressions with panel data.
If I make a OLS regression with the panel data it works without any problem. My used codes are as follows:

regress D.(viol_crime_rate asylum_pop a8_pop) i.Year [aw=adult_pop], vce(cluster panel_id)

However, if I try to make a fixed-effect panel regression, I get as a result "negative weights encountered". My used codes are as follows:

xtreg viol_crime_rate asylum_pop a8_pop i.Year [aw=avg_adult_pop], fe vce(cluster panel_id)

Does anybody know, how to solve this problem?

Thank you very much for your help.

Best
LM

[William Lisowski]

In the first regression your weights are adult_pop, in the second they are avg_adult_pop. It appears that some values of avg_adult_pop are negative, but no values of adult_pop were negative, which is why the OLS regression did not have this problem.

[Lucca Mancini]

Thank you so much William.