I'm reading a paper by Gyöngyösi and Verner (2019), and noticed that the authors have included fixed effects at the settlement (seems to be akin to village/town) and at the region level, in a number of regressions.

I do not understand why region FEs would even be required, if you are accounting for settlement varying fixed effects.

Source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3241309


1 Answer 1


page 23: "We also control for time-varying regional shocks by including fixed effects for seven major regions (Region FE)."

Hence, these "fixed-effects" seem to be equal to 1 for all observations in a given region and a given year, and zero elsewhere. I agree that the "fixed-effect" denomination may let the reader think this is fixed over time.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.