Respected members
I am currently working on my research involving panel data analysis and have come across the Panel Corrected Standard Error (PCSE) approach as a suitable method to address potential issues of heteroscedasticity and contemporaneous correlation across panels.
While I have reviewed some literature and methodological notes, I am seeking a clearer understanding of the following aspects:
1. In which situations is the PCSE approach considered more appropriate compared to alternatives such as Driscoll-Kraay or feasible GLS?
2. The practical steps for implementing PCSE in Stata specifically the conditions which we have to select amongst the mentioned conditions:
1. No autocorrelation
2. AR(1)
3. Panel special AR(1)
in applying the PCSE approach.
Thank you very much for your time and support.
With regards
Mohini