betareg

Beta Regression

Beta regression for modeling beta-distributed dependent variables on the open unit interval (0, 1), e.g., rates and proportions, see Cribari-Neto and Zeileis (2010) <doi:10.18637/jss.v034.i02>. Moreover, extended-support beta regression models can accommodate dependent variables with boundary observations at 0 and/or 1, see Kosmidis and Zeileis (2024) <doi:10.48550/arXiv.2409.07233>. For the classical beta regression model, alternative specifications are provided: Bias-corrected and bias-reduced estimation, finite mixture models, and recursive partitioning for beta regression, see GrĂ¼n, Kosmidis, and Zeileis (2012) <doi:10.18637/jss.v048.i11>.

Tests Vignettes

Available Snapshots

This version of betareg can be found in the following snapshots:

Dependencies

Imports/Depends/LinkingTo/Enhances (6)
  • R
  • flexmix
  • Formula
  • lmtest
  • modeltools
  • sandwich
  • Suggests (9)
  • bamlss
  • car
  • distributions3
  • knitr
  • numDeriv
  • partykit
  • quarto
  • statmod
  • strucchange
  • Version History