Research Article Open Access

Assessing Convergence of the Markov Chain Monte Carlo Method in Multivariate Case

Denismar Alves Nogueira1, Thelma Safadi2, Daniel Furtado Ferreira2 and Eric Batista Ferreira1
  • 1 Federal University of Alfenas, Brazil
  • 2 Federal University of Lavras, Brazil

Abstract

The formal convergence diagnosis of the Markov Chain Monte Carlo (MCMC) is made using univariate and multivariate criteria. In 1998, a multivariate extension of the univariate criterion of multiple sequences was proposed. However, due to some problems of that multivariate criterion, an alternative form of calculation was proposed in addition to the two new alternatives for multivariate convergence criteria. In this study, two models were used, one related to time series with two interventions and ARMA (2, 2) error and another related to a trivariate normal distribution, considering three different cases for the covariance matrix. In both the cases, the Gibbs sampler and the proposed criteria to monitor the convergence were used. Results revealed the proposed criteria to be adequate, besides being easy to implement.

Journal of Mathematics and Statistics
Volume 8 No. 4, 2012, 471-480

DOI: https://doi.org/10.3844/jmssp.2012.471.480

Submitted On: 8 August 2012 Published On: 8 January 2013

How to Cite: Nogueira, D. A., Safadi, T., Ferreira, D. F. & Ferreira, E. B. (2012). Assessing Convergence of the Markov Chain Monte Carlo Method in Multivariate Case. Journal of Mathematics and Statistics, 8(4), 471-480. https://doi.org/10.3844/jmssp.2012.471.480

  • 3,006 Views
  • 2,815 Downloads
  • 0 Citations

Download

Keywords

  • Convergence Criterion
  • Gibbs Sampler
  • Bayesian Inference
  • Simulation