0

I'm modelling a logistic regression with the Firth's correction, using logistf package in R. When I use the command DROP1/ADD1, R help says it is based on the penalized likelihood ratio. But is the penalization it says actually the Firth's correction or is it just a penalized likelihood?

Thanks in advance for your help.

MLavoie
  • 9,671
  • 41
  • 36
  • 56
Claudio
  • 1
  • The package description says: Firth's bias reduced logistic regression approach with penalized profile likelihood based confidence intervals for parameter estimates. So I guess the parameters are estimated with the Firth's correction, but the confidence intervals are estimated with penalized likelihood. – StatMan Aug 03 '16 at 13:39
  • i already figured it out, but thanks :) – Claudio Aug 07 '16 at 14:39

0 Answers0