Modelo de regresión para razones y proporciones: variantes y aplicaciones socio-económicas y biosanitarias

Abstract

Many researchers use regression models to analyse data and study the relationship between some variables. The linear regression model, in particular, is commonly used in applications. This paper proposes a model based on the assumption that the response is beta distributed, which is very flexible for modelling proportions. For this reason, this model is appropriate for situations where the response is restricted to the interval (0, 1). The goal is to collect the modelling and inferential procedures for beta regression, including a description of betareg package which provides the class of beta regressions in the R system for statistical computing. Furthermore, some practical applications that employ real data are mentioned and described at the end.