Hi All,

As I am replicating a paper, I will use 9 years of data instead of their 2 years. To take advantage of this fact I want to use panel data.

In this paper they use tobit and logit regression (cross section) for a dummy dependent variable to test determinants of the dummy being 1. I would like to do the same, but then with Panel Data. As I would like to incorporate both year and firm fixed effects (also industry fixed effects, but it appears these are collinear with firm fixed effects), I can't use the xtlogit or xttobit function.

What is the best way to cope with this? Is it perfectly fine to use reghdfe with a dummy variable as your dependent variable? And in addition, I seem to get very low within r-squared (0.01) when I perform the reghdfe with year and firm fixed effects absorbed, but very high overall r-squared (70%). Is this reasonable?

Best,
Martin