As we all know, if all explanatory variables are correlated with the error term, we called it the endogeneity problem and its against the Exogeneity assumption of OLS regression. I just want to know that if we generated error tern after regression running on STATA and then apply correlation analysis on all explanatory variables and the error term, should we able detect the endogeneity problem through this process or not?
Yes, you are right. If you find endogeneity you can handle the problem by using an estimator with fixed effects in a panel setting - this ismy individual strategy.In case of time series analysis you have to apply other methods. e.g. the Durbin-Wu-Hausmann-Test, unfortunately not available in Eviews.
Professor Hubert Gabrisch , Thanks a lot for your guidance, Professor can you please share the reference about your argument "If you find endogeneity you can handle the problem by using an estimator with fixed effects in a panel setting".
Again thanks for your great help :-)
I am ver sorry, that I cannot help you more. I am not an econometrician, but an economist. It my econometrical practice, it is known that in a panel setting you can mitigate the endogeneity problem, but it would not usual to make a reference - it is simply common knoledge. To help you more, I would to have a survey among the most important literture (Hausman or Heckman), read these works in detail, and that is something I would prefer to leave it to you. Very sorry!
Hubert Gabrisch , Thanks again for your kindness Professor! I really appreciate your spirit. :-)
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression