SciELO - Scientific Electronic Library Online

 
vol.46 issue1An Improved Estimator of Finite Population Variance Using two Auxiliary Variable SRSImputation of Missing Data Through Product Type Exponential Methods in Sampling Theory author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google

Share


Revista Colombiana de Estadística

Print version ISSN 0120-1751

Rev.Colomb.Estad. vol.46 no.1 Bogotá Jan./June 2023  Epub Jan 18, 2023

https://doi.org/10.15446/rce.v46nl.95989 

Original articles of research

Objective Prior Distributions to Estimate the Parameters of the Poisson-Exponential Distribution

Distribuciones previas objetivas para estimar los parámetros de la distribución Poisson-Exponencial

Fernando A. Moala1  a 

Gustavo Moraes1  b 

1 Departamento de Estatística, Faculdade de Ciências e Tecnologia, Universidade Estadual Paulista, São Paulo, Brasil


Abstract

In this paper, a set of important objective priors are examined for the Bayesian estimation of the parameters present in the Poisson-Exponential distribution PE. We derived the multivariate Jeffreys prior and the Maxi-mal Data Information Prior. Reference prior and others priors proposed in the literature are also analyzed. We show that the posterior densities resulting from these approaches are proper although the respective priors are improper. Monte Carlo simulations are used to compare the efficiencies and to assess the sensitivity of the choice of the priors, mainly for small sample sizes. This simulation study shows that the mean square error, mean bias and coverage probability of credible intervals under Gamma, Jeffreys' rule and Box & Tiao priors presented equal results, whereas Jeffreys and Reference priors showed the best results. The MDIP prior had a worse performance in all analyzed situations showing not to be indicated for Bayesian analysis of the PE distribution. A real data set is analyzed for illustrative purpose of the Bayesian approaches.

Key words: Bayesian; Poisson-Exponential; Jeffreys; MDIP; Objective; Prior

Resumen

En este artículo, se examina un conjunto de importantes priori objetivas para la estimación bayesiana de los parámetros de la distribución Poisson-Exponencial (PE). Derivamos la priori Jeffreys multivariada y la Maximal Data Information Prior. También se analizan la priori de Referencia y otras prioris propuestas en la literatura. Mostramos que las distribuciones posterioris resultantes de estos enfoques son adecuadas, aunque las respectivas prioris son impropias. Las simulaciones de Monte Carlo se utilizan para comparar las eficiencias, para evaluar la sensibilidad de la elección de las prioris, principalmente para tamaños de muestra pequeños. Este estudio de simulación muestra que los errores cuadráticos medios, el sesgo medio y la probabilidad de cobertura de los intervalos creíbles bajo la Gamma, regla de Jeffreys y Box & Tiao mostraron resultados iguales, mientras que los prioris de Jeffreys y Reference mostraron los mejores resultados. El priori MDIP tuvo un peor desempeño en todas las situaciones analizadas mostrando no estar indicado para el análisis bayesiano de la distribución PE. Se analiza un conjunto de datos reales con fines ilustrativos de los enfoques bayesianos.

Palabras clave: Bayesiano; Jeffreys; MDIP; Objetiva; Poisson-Exponencial; Priori

Full text available only in PDF format

References

Belaghi, R., Asi, M., Alma, O., Singh, S. & Vasfi, M. (2019), 'Estimation and prediction for the poisson-exponential distribution based on type-ii censored data', American Journal of Mathematical and Management Sciences 38(1), 961-15. [ Links ]

Berger, J. O. & Bernardo, J. M. (1992), 'On the development of the reference prior method', Bayesian Statistics 4(4), 35-60. [ Links ]

Bernardo, J. M. (1979), 'Reference posterior distributions for bayesian inference', Journal Roy al Statistical Society 41(2), 113-147. [ Links ]

Box, G. E. P. & Tiao, G. C. (1973), Bayesian inference in statistical analysis. [ Links ]

Cancho, V. G., Louzada-Neto, F. & Barriga, G. D. C. (2011), 'The poisson-exponential lifetime distribution', Computational Statistics and Data Analysis 55, 677-686. [ Links ]

Jeffreys, S. H. (1967), Theory of probability, number 3, London, Oxford U. Press. [ Links ]

Lawless, J. F. (2003), Statistical Models and Methods for Lifetime Data, Vol. 2 edition, Wiley, New York. [ Links ]

Louzada-Neto, F., Cancho, V. G. & Barriga, G. D. C. (2011), 'The poisson-exponential distribution: a bayesian approach', Journal of Applied Statistics 38(6), 1239-1248. [ Links ]

Rodrigues, G., Louzada, F. & Ramos, P. (2018), 'Poisson-exponential distribution: different methods of estimation', Journal of Applied Statistics 45(1), 128-144. [ Links ]

Singh, S., Singh, U. & Kumar, M. (2014), 'Estimation for the parameter of poisson-exponential distribution under bayesian paradigm', Journal of Data Science 12, 157-173. [ Links ]

Tibshirani, R. (1987), 'Noninformative priors for one parameters of many', Biometrika 76, 604-608. [ Links ]

Tomazella, V. L. D., Cancho, V. G. & Louzada-Neto, F. (2013), 'Bayesian reference analysis for the poisson-exponential lifetime distribution', Chilean Journal of Statistics 4(1), 99-113. [ Links ]

Zellner, A. (1977), Maximal data information prior distributions, in A. Aykac & C. Brumat, eds, 'New methods in the applications of bayesian methods', North-Holland, Amsterdam. [ Links ]

Zellner, A. (1984), 'Maximal data information prior distributions', Basic Issues in Econometrics. [ Links ]

Zellner, A. (1990), Bayesian methods and entropy in economics and econometrics, in J. W. T. Grandy & L. H. Schick, eds, 'Máximum Entropy and Bayesian Methods', Dordrecht, Netherlands: Kluwer Academic Publishers, pp. 17-31. [ Links ]

Zellner, A. (1996), 'Models, prior information and bayesian analysis', Journal of Econometrics 75(1), 51-68. [ Links ]

Zellner, A. & Min, C.-K. (1992), Bayesian analysis, model selection and prediction, University of Chicago, Gradúate School of Business, Department of Economics. [ Links ]

Received: May 2022; Accepted: November 2022

a Professor. e-mail: femoala@fct.unesp.br

b Professor. e-mail: gustavo.perez@unesp.br

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License