Nice video thank you for sharing. I would like to ask you if I have panel data with heteroscedasticity, autocorrelation and cross-sectional dependence and I try to find the appropriatr model, I have to choose between GLS and PCSE, what can I do, noting that the N is greater than the T
@@Paneldataanalysis-go1ky Thank you for your reply. I am confused a little because I find many videos day that I go for PCSE as in in the paper titled (Robust standard errors for pane regressions with cross section dependence) when T is greater than N I use GLS, but no mention to my case where N is greater than T. I hope you can help me to find an evidenced -based way to settle this problem. Thank you so much.
Nice video thank you for sharing. I would like to ask you if I have panel data with heteroscedasticity, autocorrelation and cross-sectional dependence and I try to find the appropriatr model, I have to choose between GLS and PCSE, what can I do, noting that the N is greater than the T
I will go for GLS with the panels(correlated) option.
@@Paneldataanalysis-go1ky Thank you for your reply. I am confused a little because I find many videos day that I go for PCSE as in in the paper titled (Robust standard errors for pane regressions with cross section dependence) when T is greater than N I use GLS, but no mention to my case where N is greater than T. I hope you can help me to find an evidenced -based way to settle this problem. Thank you so much.
@@usamaahmed6056 Yes, that's right. xtgls command is used when T > N. For N>T, xtgee command is appropriate.
@@Paneldataanalysis-go1ky Thank you so much for your reply