Fixed effects and wald test
Posted: Thu Mar 07, 2019 2:54 am
Hello,
I have set of data where i perforemd the hausman-test to check if fixed or random effetc should be applied.
With a very low p value of 0.000009 it suggets to take fixed effetcs.
The problem if i tried a fixed effects equation with cross-section set to fix and than want to choose the option GLS weights i get an error : "positive or non-negative argument to function expected in computation of group weight (variance) "
does anybody have an idea whats the error here?
I have set of data where i perforemd the hausman-test to check if fixed or random effetc should be applied.
With a very low p value of 0.000009 it suggets to take fixed effetcs.
The problem if i tried a fixed effects equation with cross-section set to fix and than want to choose the option GLS weights i get an error : "positive or non-negative argument to function expected in computation of group weight (variance) "
does anybody have an idea whats the error here?