I have heteroskedastic and serrialy correlated (autocorrelated) panel data. I want to test it with both dynamic and static models. For the dynamic test, I use GMM and the results of GMM is parallel to results of Newey-West estimator. Newey-West estimator is OLS estimator that robust disturbances heteroscedastic and autocorrelated of type MA(q)^2. But, whenever I test the model with fixed effect regression, I find completely different results.

How can I understand what type of the autocorrelation my data has and which estimator should I use? I don't understand the differences between these estimators and below commands in terms of application.

enter image description here


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.