This chapter aims to study the performances of a new regression model for continuous variables with bounded support that extends the well-known beta regression model. Under the new regression model, the response variable is assumed to have a flexible beta (FB) distribution, a special mixture of two beta distributions that can be interpreted as the univariate version of the flexible Dirichlet distribution. The chapter introduces the FB distribution, proposes a reparameterization that is designed for this regression context, and enables a very clear interpretation of the new parameters. It defines the FB regression (FBR) model and interprets it as mixture of regression models. The chapter provides details concerning Bayesian inference and the Gibbs sampling algorithm specifically designed for mixture models. It performs an illustrative application on a real data set in order to evaluate the performance of the FBR model and compare it with the BR and beta regression ones.

Migliorati, S., Di Brisco, A., Ongaro, A. (2019). The Flexible Beta Regression Model. In C.H. Skiadas, J.R. Bozeman (a cura di), Data Analysis and Applications 1 (pp. 39-52). ISTE [10.1002/9781119597568.ch3].

The Flexible Beta Regression Model

Migliorati, S;Di Brisco, AM
;
Ongaro, A
2019

Abstract

This chapter aims to study the performances of a new regression model for continuous variables with bounded support that extends the well-known beta regression model. Under the new regression model, the response variable is assumed to have a flexible beta (FB) distribution, a special mixture of two beta distributions that can be interpreted as the univariate version of the flexible Dirichlet distribution. The chapter introduces the FB distribution, proposes a reparameterization that is designed for this regression context, and enables a very clear interpretation of the new parameters. It defines the FB regression (FBR) model and interprets it as mixture of regression models. The chapter provides details concerning Bayesian inference and the Gibbs sampling algorithm specifically designed for mixture models. It performs an illustrative application on a real data set in order to evaluate the performance of the FBR model and compare it with the BR and beta regression ones.
Capitolo o saggio
proportions, beta regression, mixture models, MCMC
English
Data Analysis and Applications 1
Skiadas, CH; Bozeman, JR
feb-2019
2019
9781786303820
ISTE
39
52
Migliorati, S., Di Brisco, A., Ongaro, A. (2019). The Flexible Beta Regression Model. In C.H. Skiadas, J.R. Bozeman (a cura di), Data Analysis and Applications 1 (pp. 39-52). ISTE [10.1002/9781119597568.ch3].
none
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/220537
Citazioni
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
Social impact