I'm trying to translate Stata results into R and with the existing methods, I can only recreate standard errors up the the 5th digit after which they disagree (see a blogpost on that issue here: https://blog.theleapjournal.org/2016...rd-errors.html).

Now I could of course build covariance matrices manually and should be able to reproduce Stata (no problem in R of course) but I'm failing to find the exact small sample adjustment it uses. Is this recorded somewhere? I'm interested in a regression of this form:

xtreg y x, fe vce(cluster individual)

Any hint where to look would be appreciated!