Beta Regression

This post focuses on one of the more curious models in contemporary statistics, a specification for proportions that is either called fractional logit or quasi-Binomial. An earlier version of this blog post had a much more negative take on the fractional logit specification.

Twitter (or what’s left of it) was recently ablaze with a discussion of two smart working papers, one by Jiafeng Chen and Jonathan Roth and the other by John Mullahy and Edward Norton.

Introduction Limited dependent variables, or continuous variables with lower and upper bounds, are quite common in the social sciences but do not fit easily with existing statistical models. In this Rmarkdown document, I show why these issues are important to consider when modeling your data, discuss existing R packages useful for fitting these models, and also present ordbetareg, an R package with a new variant of Beta regression that builds on and simplifies existing approaches (see paper here that is forthcoming in Political Analysis).