Ordered Beta Regression: A Parsimonious, Well-Fitting Model for Continuous Data with Lower and Upper Bounds

Abstract I propose a new model, ordered Beta regression, for continuous distributions with both lower and upper bounds, such as data arising from survey slider scales, visual analog dose–response relationships. This model employs the cut point technique popularized by logit to fit single linear (0,1) degenerate [0,1] responses. The can be estimated or without observations at is general solution these types of data. Employing Monte Carlo simulation, show that noticeably more efficient than ordinary least squares zero-and-one-inflated rescaled fractional while fully capturing nuances in outcome. apply replication Aidt Jensen (2014, European Economic Review 72, 52–75) study suffrage extensions Europe. R package ordbetareg facilitate hierarchical, dynamic, multivariate modeling.