Simultaneous Modeling of Mean and Variance In Augmented Mixed Beta Regression

Document Type : Original Paper

Authors

1 Statistical Research and Training Center, Tehran, Iran

2 Department of statistics, Islamic Azad University Tehran North Branch, Tehran, Iran

Abstract

Augmented Beta Regression models are used for modeling data such as rate, ratio or percentage. This model is made by combining the Beta distribution on the interval (0,1) and two degenerate distributions at 0 and 1. By reparameterizing the beta distribution, the mean and precision parameters are modeled with a structure including fixed and random effects. In this paper, simultaneous modeling of mean and precision the augmented mixed beta regression models is presented and the model efficiency in simulation studies by Bayesian approach is investigated. Next, the application of this model to analyze the proportions of employed persons in every household based on the results of the Statistical Center of Iran is shown and at the end, conclusion and results are presented.

Keywords

Main Subjects


[1] نتایج آمارگیری نیروی کار ، ( ١٣٩٢ )، تهران، مرکز آمار ایران.
[2] Branscum, A.J., Johnson, W.O. and Thurmond, M. (2007), Baysian Beta Regression Applications to
Houshold Expenditure Data and Genetic Distance Between Food and Mouth Dieseas Viruses, Australian
& New Zealand Journal of Statistics, 49, 287-301.
[3] Bonat, W.H., Ribeiro, P.J. and Zeviani, W.M. (2013), Likelihood Analysis for a Class of Beta Mixed,
Models Cornell University Library, arXiv Preprint arXiv: 1312.2413.
[4] Cepeda, E. D., and Gamerman, D. (2005), Bayesian Methodology for Modeling Parameters in the Two
Parameter ExponentialF amily, Revista Estadística, 57, 168-169.
[5] Cepeda, E. .D, Migon, H. S., Garrido, L. and Achcar, J. A. (2014), Generalized Linear Models with
Random Effects in the Two-ParameterE xponentialF amily, Journal of Statistical Computation and
Simulation, 84, 513-525.
[6] Carlin, B. P. and Louis, T. A. (2008), Bayesian Methods for Data Analysis, Mineapolis, CRC Press.
[7] Ferrari, S. and Cribari, F. (2004), Beta Regression for Modelling Rates and Proportions, Journal of
Applied Statistics,31, 799-815.
[8] Figueroa-Zúñiga, J. I., Arellano-Valle, R. B. and Ferrari, S. L. (2013), Mixed Beta Regression: A
Bayesian Perspective, Computational Statistics & Data Analysis,61, 137– 147.
[9] Fallah Mohsenkhani, Z., Mohammadzadeh, M. and Baghfalaki, T. (2019), Augmented Mixed Beta
Regression Models with Skew-Normal Independent Distributions: Bayesian Analysis of Labor Force
Data, Communications in Statistics-Simulation and Computation,Volume 48, Issue 7, 2147-2164.
[10] Fong, Y., Rue, H. and Wakefield, J. (2010), Bayesian Inference for Generalized Linear Mixed Models,
Biostatistics, 11, 397-412.
[11] Galvis, M. D., Dipankar, B. and Victor, H. L. (2014), Augmented Mixed Beta Regression Models for
Periodontal Proportion Data, Preprinted, (In Press), Statistics in Medicine.
[12] Gelman, A., Rubin, D. B., (1992). Inference from Iterative Simulation Using Multiple Sequences,
Statistical Science7, 457–511.
[13] Gelfand, A. E. and Dey, D. K. (1994), Bayesian Model Choice: Asymptotics and Exact Calculations,
Journal of the Royal Statistical Society, Series B (Methodological), 501-514.
[14] Heidelberger, P. and Welch, P. D. (1981), A Spectral Method for Confidence Interval Generation and
Run Length Control in Simulations, Communications of the ACM, 24, 233-245.
[15] Nogarotto, D, Azevedo, C, Bazan, J.Bayesian (2020), modeling and prior sensitivity analysis for
zero–one augmented beta regression models with an application to psychometric data, Brazilian Journal
of Probability and Statistics, 304-322.
[16] Ospina, R. and Ferrari, S. L. (2010), Inflated Beta Distributions, Statistical Papers. 51,111- 126.
[17] Paolino, P. (2001), Maximum Likelihood Estimation of Models with Beta-Distributed Dependent
Variables, Political Analysis, 9, 325-346.
[18] Parker, A, Bandyopadhyay, D and Slate, E. (2014), A spatial augmented beta regression model for
periodontal proportion data, Statistical Modelling, vol. 14, 503-521.
[19] Smithson, M. and Verkuilen, J. (2006), A Better Lemon Squeezer? Maximum-Likelihood Regression
with Beta-Distributed Dependent Variables, Psychological Methods, 11, 54-71.
[20] Verkuilen, J. and Smithson, M. (2012), Mixed and Mixture Regression Models for Continuous
Bounded Responses Using the Beta Distribution, Journal of Educational and Behavioral Statistics,
37,82-113.
[21] Zimprich, D. (2010), Modeling Change in Skewed Variables Using Mixed Beta Regression Models,
Research in Human Development, 7, 9-26.
Volume 12, Issue 1
April 2022
Pages 13-23
  • Receive Date: 09 August 2020
  • Revise Date: 18 October 2021
  • Accept Date: 20 November 2021
  • First Publish Date: 26 January 2022