I hope I'm not barking up the wrong tree posting here. I'm trying to do a LR test of two different regressions (with vs w/o interaction terms) where the dependent variable is continuous and bounded by [0,1]. When the model is a beta regression (using betareg), the R output includes the chi squared quantile and p-value But, when I do a fractional logit regression (glm with a quasibinomial link function) I don't get those terms in the output. Am I doing something wrong or is this an inherent feature of fractional logit models?
Reproducible code below. Thank you!
test = data.frame(y = rbeta(1000,1,2), x1 = rnorm(1000), x2 = rnorm(1000), x3 = rnorm(1000))
betar1 = betareg::betareg(y ~ x1 + x2 + x3, data = test)
betar2 = betareg::betareg(y ~ x1 + x2 + x1:x2 + x3, data = test)
fracr1 = glm(y ~ x1 + x2 + x3, family = quasibinomial("logit"), data = test)
fracr2 = glm(y ~ x1 + x2 + x1:x2 + x3, family = quasibinomial("logit"), data = test)