Whats the difference between logistic regression and fractional response model?

Question

1. Can anyone tell me the theory behind fractional response model, how it really works?

2. I wonder if the logistic regression works only with binary variable {0,1}, why when conducting a GLM with quasibinomial family on fractional variable [0-1], it gave me same results in the estimates as the fractional response model?

Answer 1

I have not seen the term fractional response model before, but assuming it is models for fractions between zero and one as response, as it is used in this paper, it seems to be a term used by econometricians for logistic quasi-likelihood.

If you use R, these models are estimated simply by switching from the binomial family function to the quasibinomial family function.