betaregscale: Beta Regression for Interval-Censored Scale-Derived Outcomes

Maximum-likelihood estimation of beta regression models for responses derived from bounded rating scales. Observations are treated as interval-censored on (0, 1) after a scale-to-unit transformation. The complete likelihood supports mixed censoring types: uncensored (exact), left-censored, right-censored, and interval-censored observations. Both fixed- and variable-dispersion submodels are supported, with flexible link functions for the mean and precision components. A compiled C++ backend (via Rcpp and RcppArmadillo) provides numerically stable, high-performance log-likelihood evaluation. Standard S3 methods (print(), summary(), coef(), fitted(), residuals(), predict(), plot(), confint(), vcov(), logLik(), AIC(), BIC()) are available for fitted objects.

Main functions

brs: Unified fitting interface for both fixed- and variable-dispersion models.
brs_fit_fixed: Fit a fixed-dispersion model.
brs_fit_var: Fit a variable-dispersion model.
brsmm: Fit a mixed-effects beta interval model with Gaussian random intercepts.
brs_sim: Simulate interval-censored data from fixed or variable-dispersion beta models.
brs_bootstrap: Parametric bootstrap confidence intervals for brs model parameters.
brs_cens: Visual and tabular summary of censoring structure.
brs_prep: Pre-process analyst data (validate, classify censoring, and rescale) before model fitting.

S3 methods

Objects of class "brs" support: print(), summary(), coef(), vcov(), logLik(), AIC(), BIC(), nobs(), formula(), model.matrix(), fitted(), residuals(), predict(), confint(), and plot().

The coef() and vcov() methods accept a model = c("full", "mean", "precision") argument following the betareg package convention.

Censoring types

The complete likelihood supports four censoring types, classified automatically by brs_check:

\(\delta = 0\) (exact): Continuous observations in (0, 1).
\(\delta = 1\) (left-censored): Observations at the scale minimum (y = 0).
\(\delta = 2\) (right-censored): Observations at the scale maximum (y = ncuts).
\(\delta = 3\) (interval-censored): Standard scale observations between the boundaries.

References

Lopes, J. E. (2023). Modelos de regressao beta para dados de escala. Master's dissertation, Universidade Federal do Parana, Curitiba. URI: https://hdl.handle.net/1884/86624.

Ferrari, S. L. P., and Cribari-Neto, F. (2004). Beta regression for modelling rates and proportions. Journal of Applied Statistics, 31(7), 799–815. doi:10.1080/0266476042000214501

Author

Maintainer: José Evandeilton Lopes evandeilton@gmail.com (ORCID)

Authors:

Wagner Hugo Bonat (ORCID)