To Bridge, to Warp or to Wrap? A Comperative Study of Monte Carlo Methods for Efficient Evaluation of Marginal Likelihood
Computational Statistics & Data Analysis, Vol. 56, No. 11, pp. 3398-3414, 2012
TI Discussion Paper No. 09-017/4
44 Pages Posted: 26 Feb 2009 Last revised: 13 May 2015
Date Written: February 26, 2009
Abstract
Important choices for efficient and accurate evaluation of marginal likelihoods by means of Monte Carlo simulation methods are studied for the case of highly non-elliptical posterior distributions. We focus on the situation where one makes use of importance sampling or the independence chain Metropolis-Hastings algorithm for posterior analysis. A comparative analysis is presented of possible advantages and limitations of different simulation techniques; of possible choices of candidate distributions and choices of target or warped target distributions; and finally of numerical standard errors. The importance of a robust and flexible estimation strategy is demonstrated where the complete posterior distribution is explored. In this respect, the adaptive mixture of Student-t distributions of Hoogerheide et al.(2007) works particularly well. Given an appropriately yet quickly tuned candidate, straightforward importance sampling provides the most efficient estimator of the marginal likelihood in the cases investigated in this paper, which include a non-linear regression model of Ritter and Tanner (1992) and a conditional normal distribution of Gelman and Meng (1991). A poor choice of candidate density may lead to a huge loss of efficiency where the numerical standard error may be highly unreliable.
Keywords: marginal likelihood, Bayes factor, importance sampling, Markov chain Monte Carlo, bridge sampling, adaptive mixture of Student-t distributions
JEL Classification: C11, C15, C52
Suggested Citation: Suggested Citation
Do you have negative results from your research you’d like to share?
Recommended Papers
-
By Lennart F. Hoogerheide, Johan F. Kaashoek, ...
-
Simulation Based Bayesian Econometric Inference: Principles and Some Recent Computational Advances
By Lennart F. Hoogerheide, H. K. Van Dijk, ...
-
Bayesian Forecasting of Value at Risk and Expected Shortfall Using Adaptive Importance Sampling
-
AdMit: Adaptive Mixtures of Student-t Distributions
By David Ardia, Lennart F. Hoogerheide, ...
-
By David Ardia, Lennart F. Hoogerheide, ...
-
By Moshe Leshno, Valdimir Ya. Lin, ...
-
By Moshe Leshno and Shimon Schocken
-
On the Practice of Bayesian Inference in Basic Economic Time Series Models Using Gibbs Sampling
By Michiel De Pooter, Rene Segers, ...