[Clyde Schechter]

I'm not familiar with -regsave-. In any case, you don't need a special program for this.

xtreg Z_score LLR LLP NPA i.year, fe vce(cluster id)
gen r2_full = e(r2_w)
xtreg i.year, fe vce(cluster id)
gen r2_reduced = e(r2_w)
gen r2_diff = r2_full'- r2_reduced

will put the R²'s from both models, and the difference into variables in your data set as new "variables." Of course, they are constant, so the wisdom of saving them as variables is questionable--but depending on your other plans for this it might make sense to do that.

Actually I have four bank risk proxies (Z_score, NPA, LLR, LLP) and for each I would like to calculate the R².

I don't understand what you mean by this and the rest of #16.

[Katharina Maier]

It worked perfectly and was exactly what I was looking for! Thank you so much!