Multi level fixed effects with binary dependent variable

[Anna Bonetto]

Dear Statalisters,

I would like to use the REGHDFE command with a binary dependent variable and a binary independent variable with many levels of fixed effects, but I don't know if I can interpret the magnitude and the significance of the coefficient as in the linear case. If not, which command I should use?

[Phil Bromiley]

Look at the multilevel mixed effects set of routines. Some do use regression for binary dv's in which case they interpret the predicted value as sort of like a predicted probability. However, as has been discussed elsewhere, this has serious drawbacks.