I have a question regarding the xtgls command and heteroskedasticity. I
have seen that several other people have asked similar questions, but I
couldn't find any answers.
I am estimating a random effects model and I know that you can do a LR test
for heteroskedasticity using xtgls, igls. That's what I have done and
though I got some output indicating heteroskedasticity, the iteration is
actually never achieved. Does that mean there is a problem with my data (I
have only 69 observations and 8 years) and what does that mean for the
goodness of the LR test?
Moreover, if I do any xtgls regression, how do I get an r squared /
adjusted r squared?
