Page 1 of 1

Explaining a proportion

Posted: Fri Mar 09, 2012 7:18 am
by Philippe Rous
Dear all,

I want to build up a model explaining a variable Y which can take any value between 0 and 1 (it may be a proportion or an efficiency score...). My problem is quite different from the one for which Y would be a truncated variable where Y could take some unobservale value out of [0, 1] range. Here, Y cannot take any value (even unobservable) out of [0, 1] range. Is TOBIT a suitable choice ?

Is the model Y = F(Xb + e) where F(.) is a CDF a better choice

Thanks a lot for your help !

Best regards

Phil