I am working on a Dynamic Panel Model in Stata, i tested for the serial correlation and the heteroskedasticity (LM test and Breusch-Pagan test), and there is no serial correlation but the panels are heteroskedastic.
So, my first question is: how do i do to correct the heteroskedasticity ??
Please note that I'm working on a Dynamic Panel ( ARDL Panel ). I tested for the suitable model (MG or PMG or DFE) and the PMG is the suitable model for my Data according to the test.
My second question: is the heteroscedasticity a serious problem in an ARDL Panel ??