I would like to use estimations of "fixed effect" from my panel data model as a dependent variable and study whether this depends on time invariant characteristics. I've not been able to find an article, text-book, etc., using this methodology which makes me think that there might be something precluding one to make such estimation.

If you have some idea on why this is not feasible, or instead, you know some article treating this I would really appreciate if you let me know.

Thanks in advance.

  • $\begingroup$ There was essentially the same question asked here: stats.stackexchange.com/q/100168/241093, and one of the answers provides some information that may be helpful to you. There is also a link in the comment to the question to a paper that apparently illustrates this method. $\endgroup$
    – AlexK
    May 19 '19 at 9:19