Dynamic Panel model with binary outcome variable

[Luis Mijares Castaneda]

Hello, I was wondering if there was a version of the command -xtdpdqml- that could incorporate logistic regression. I was interested in using a binary outcome variable. Or If there was another method of modeling lagged dependent variables with binary outcomes.

[Sebastian Kripfganz]

You can have a look at the community-contributed command logitfe and the associated literature on bias-corrected estimation of nonlinear dynamic panel data models (see the logitfe help file). Alternatively, you can use xtprobit with the Wooldridge (2005) approach to the initial-conditions problem.

[Stephen Jenkins]

Re Sebastian's second suggestion: I tihnk this is packaged in -xtpdyn-, accompanying article in Stata Journal, volume 18, number 4: st0543. Use -search xtpdyn- to find it

[Jeff Wooldridge]

Luis contacted me by email, and I suggested the method -- along with Stata code -- from my 2005 Journal of Applied Econometrics paper, as linked by Sebastian.

[George Ford]

Also see: https://www.statalist.org/forums/for...ridge-approach