Dear all,

I am using reghdfe and have few clusters. Throughout, I Wild-Cluster bottstrap my p-values. However, I noticed that the inference in the reghdfe is already quite conservative. For instance, a t-value of 2.11 results in a p-value of 0.052. My educated guess is that reghdfe uses G-1 degrees of freedom for the t-distribution, with G being the number of clusters. Am I correct? I am totally okay with that. I just want to know what is happening. Can someone point me into the right direction?

example code:

reghdfe y x, absorb(fe) cluster(clusterdimension)

Thank you very much.

Daniel