Published
MCMC Methods for Sample Generation from a New Bivariate Distribution
Métodos MCMC para generar muestras de una nueva distribución bivariada
DOI:
https://doi.org/10.15446/rce.v48n2.117153Keywords:
Adaptive random walk, Bivariate probability distribution, Convergence diagnostics, Gibbs sampling, MCMC, Metropolis-Hastings. (en)Caminata aleatoria adaptativa, Diagnósticos de convergencia, Distribución de probabilidad bivariada, Gibbs sampling, MCMC, Metropolis-Hastings. (es)
Downloads
This article introduces a novel bivariate probability distribution derived through a transformation-based approach, along with the closed-form expression of its l-th order joint moment. Although the distribution may be employed as a prior for the shape parameters of the Beta distribution, the main focus of this work lies in evaluating the convergence behavior of Markov Chain Monte Carlo (MCMC) algorithms designed to generate samples from this new distribution. A simulation strategy is analyzed, consisting of a Gibbs sampling scheme in which an adaptive random walk Metropolis-Hastings (ARWMH) algorithm is used to sample from one of the full conditional distributions, employing a four-parameter Beta distribution as the proposal. Convergence is assessed using diagnostics such as the effective sample size (ESS) and the potential scale reduction factor (R-hat). The results show that when the elements of the parameter vector ϕ differ, the empirical moments obtained from the chains approximate the theoretical values accurately. However, when all components of ϕ are equal, the estimates of variance and covariance deviate considerably, revealing a sensitivity to symmetry in the geometry of the new distribution. A brief application in a Bayesian context is also presented, in which the new distribution is used as a prior and the Beta distribution as the likelihood. This application confirms that the proposed sampling methods yield empirical moments that are consistent with the theoretical ones, thus supporting the validity of the strategy. The contributions of this study are relevant to the design, evaluation, and implementation of MCMC techniques for sampling from complex distributions in Bayesian inference.
Este artículo presenta una nueva distribución de probabilidad bivariada construida mediante una estrategia basada en transformaciones, junto con la expresión cerrada de su momento conjunto de orden l. Aunque esta distribución puede utilizarse como distribución a priori para los parámetros de forma de la distribución Beta, el enfoque principal del trabajo se centra en evaluar el comportamiento de convergencia de algoritmos Monte Carlo Cadenas de Markov (MCMC) diseñados para generar muestras aleatorias de dicha distribución. Se analiza una estrategia de simulación que consiste en un esquema de muestreo de Gibbs, en el cual se emplea un algoritmo Metropolis-Hastings con caminata aleatoria adaptativa (MHCAA) para generar muestras de una de las distribuciones condicionales completas, utilizando como propuesta una distribución Beta de cuatro parámetros. La convergencia se evalúa mediante diagnósticos como el tamaño efectivo de muestra (ESS) y el factor de reducción de escala potencial (R-hat). Los resultados muestran que, cuando los componentes del vector de parámetros ϕ son diferentes, los momentos empíricos obtenidos aproximan adecuadamente los valores teóricos. En contraste, cuando todos los componentes de ϕ son iguales, se evidencian desviaciones significativas en las estimaciones de la varianza y la covarianza, lo que sugiere sensibilidad ante configuraciones simétricas de la nueva distribución. Se presenta además una aplicación breve en el contexto bayesiano, usando la nueva distribución como a priori y la distribución Beta como verosimilitud, confirmando que las muestras generadas reproducen momentos a priori empíricos coherentes con los valores teóricos. Las contribuciones del estudio son relevantes para el diseño, evaluación e implementación de técnicas MCMC en contextos de inferencia bayesiana con distribuciones complejas.
References
Althubyani, F. A., El-Bar, A. M. T. A., Fawzy, M. A. & Gemeay, A. M. (2022), 'A new 3-parameter bounded beta distribution: Properties, estimation, and applications', Axioms 11(10), 504. DOI: https://doi.org/10.3390/axioms11100504
Arroyo, L. G., Lasso, F. A. & Tovar, J. R. (2022), 'Propuesta para obtener distribuciones previas para los parámetros de la distribución beta', Investigación Operacional 43(1), 51-63.
Bouguila, N., Ziou, D. & Monga, E. (2006), 'Practical bayesian estimation of a finite beta mixture through gibbs sampling and its applications', Statistics and Computing 16(2), 215-225. DOI: https://doi.org/10.1007/s11222-006-8451-7
Bran-Cardona, P. A., Orozco-Castañeda, J. M. & Nagar, D. K. (2011), 'Generalización bivariada de la distribución kummer-beta', Revista Colombiana de Estadística 34(3), 497-512.
Defiatri, D. & Damayanti, R. W. (2023), Acceptance sampling mil-std 105e for quality control: A case study, in 'E3S Web of Conferences', Vol. 465, EDP Sciences, p. 02043. DOI: https://doi.org/10.1051/e3sconf/202346502043
Kaminskiy, M. P. & Krivtsov, V. V. (2005), 'A simple procedure for bayesian estimation of the weibull distribution', IEEE Transactions on Reliability 54(4), 612-616. DOI: https://doi.org/10.1109/TR.2005.858093
Ma, Z. & Leijon, A. (2010), Expectation propagation for estimating the parameters of the beta distribution, in '2010 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)', IEEE, pp. 2082-2085. DOI: https://doi.org/10.1109/ICASSP.2010.5495085
Ma, Z. & Leijon, A. (2011), 'Bayesian estimation of beta mixture models with variational inference', IEEE Transactions on Pattern Analysis and Machine Intelligence 33(11), 2160-2173. DOI: https://doi.org/10.1109/TPAMI.2011.63
Mandouh, R. M. & Muhammed, H. Z. (2024), 'On a bivariate bounded distribution: Properties and estimation', Computational Journal of Mathematical and Statistical Sciences 3(1), 125-144. DOI: https://doi.org/10.21608/cjmss.2023.251099.1029
Martínez-Flórez, G., Vergara-Cardozo, S., Tovar-Falón, R. & Rodriguez-Quevedo, L. (2023), 'The multivariate skewed log-birnbaum-saunders distribution and its associated regression model', Mathematics 11(5), 1095. DOI: https://doi.org/10.3390/math11051095
Moschen, L. M. & Carvalho, L. M. (2023), 'Bivariate beta distribution: Parameter inference and diagnostics', arXiv preprint arXiv:2303.01271. https://arxiv.org/abs/2303.01271.
Robert, C. P. & Casella, G. (1999), Monte Carlo Statistical Methods, 1st edn, Springer. DOI: https://doi.org/10.1007/978-1-4757-3071-5_1
Robert, C. P. & Casella, G. (2010), Introducing Monte Carlo Methods with R, Use R!, Springer. DOI: https://doi.org/10.1007/978-1-4419-1576-4
Tovar, J. R. (2012), 'Eliciting beta prior distributions for binomial sampling', Revista Brasileira de Biometria 30(1), 159-172.
How to Cite
APA
ACM
ACS
ABNT
Chicago
Harvard
IEEE
MLA
Turabian
Vancouver
Download Citation
License

This work is licensed under a Creative Commons Attribution 4.0 International License.
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).






