Adaptive proposal construction for reversible jump MCMC. (English) Zbl 1190.62048
Bayesian inference for model selection is considered. The technique of Markov chain Monte Carlo posterior simulations is discussed for this problem, in which a locally adaptive updating scheme is constructed for the Metropolis-Hastings algorithm. It is based on the idea that for good proposal distributions the acceptance ratio should be approximately constant, and so it’s derivatives should be nearly zero. The general scheme is applied to order selection for autoregressive models. The performance of the algorithm is assessed by numerical examples.
Reviewer: R. E. Maiboroda (Kyïv)
MSC:
| 62F15 | Bayesian inference |
| 65C40 | Numerical analysis or methods applied to Markov chains |
| 62M10 | Time series, auto-correlation, regression, etc. in statistics (GARCH) |
Keywords:
model selection; Markov chain Monte-Carlo; Metropolis-Hastings algorithm; random number generation; autoregressionSoftware:
BUGSReferences:
| [1] | Barbieri, A reversible jump MCMC sampler for Bayesian analysis of ARMA time series (1996) |
| [2] | Brooks, Classical model selection via simulated annealing, J. Roy. Statist. Soc. Ser. B Stat. Methodol. 65 pp 503– (2003) · Zbl 1065.62051 |
| [3] | Brooks, Efficient construction of reversible jump MCMC proposal distributions (with discussion), J. Roy. Statist. Soc. Ser. B Stat. Methodol. 65 pp 1– (2003) · Zbl 1063.62120 · doi:10.1111/1467-9868.03711 |
| [4] | Carlin, Bayesian model choice via Markov chain Monte Carlo, J. Roy. Statist. Soc. Ser. B Stat. Methodol. 57 pp 473– (1995) · Zbl 0827.62027 |
| [5] | Chib, Bayes inference in regression models with ARMA(p,q) errors, J. Econometrics 64 pp 183– (1994) · Zbl 0807.62065 |
| [6] | Gåsemyr, On an adaptive version of the Metropolis-Hastings algorithm with independent proposal distribution, Scand. J. Statist. 30 pp 159– (2003) |
| [7] | Gåsemyr, An application of adaptive independent chain Metropolis-Hastings algorithms in Bayesian hazard rate estimation, Methodol. Comput. Appl. Probab. 6 pp 293– (2004) · Zbl 1049.65010 |
| [8] | Gilks, Adaptive direction sampling, The Statistician 43 pp 179– (1994) |
| [9] | Gilks, Adaptive Markov chain Monte Carlo through regeneration, J. Amer. Statist. Assoc. 93 pp 1045– (1998) · Zbl 1064.65503 |
| [10] | Godsill, On the relationship between MCMC model uncertainty methods, J. Comput. Graph. Statist. 10 pp 230– (2001) |
| [11] | Green, Reversible jump MCMC computation and Bayesian model determination, Biometrika 82 pp 711– (1995) |
| [12] | Haario, Adaptive proposal distribution for random walk Metropolis algorithm, Comput. Statist. 14 pp 375– (1999) · Zbl 0941.62036 |
| [13] | Haario, An adaptive Metropolis algorithm, Bernoulli 7 pp 223– (2001) · Zbl 0989.65004 |
| [14] | Hastings, Monte Carlo sampling methods using Markov chains and their applications, Biometrika 57 pp 97– (1970) · Zbl 0219.65008 |
| [15] | Huerta, Priors and component structures in autoregressive time series models, J. Roy. Statist. Soc. Ser. B Stat. Methodol. 51 pp 881– (1999) · Zbl 0940.62079 |
| [16] | Kass, Bayes factors, J. Amer. Statist. Assoc. 90 pp 773– (1995) |
| [17] | Ntzoufras, Bayesian variable and link determination for generalised linear models, J. Statist. Plann. Inference 111 pp 165– (2003) · Zbl 1033.62026 |
| [18] | Priestley, Spectral analysis and time series (1981) · Zbl 0537.62075 |
| [19] | Ripley, Stochastic simulation (1987) · Zbl 0613.65006 |
| [20] | Sahu, Self regenerative Markov chain Monte Carlo with adaptation, Bernoulli 9 pp 395– (2003) · Zbl 1044.62033 |
| [21] | Sargent, Structured Markov chain Monte Carlo (1998) |
| [22] | Spiegelhalter, BUGS: Bayesian inference using Gibbs sampling (1996) |
| [23] | Stephens, Bayesian analysis of mixture models with an unknown number of components - an alternative to reversible jump methods, Ann. Statist. 28 pp 40– (2000) · Zbl 1106.62316 |
| [24] | Tierney, Some adaptive Monte Carlo methods for Bayesian inference (1999) |
| [25] | Troughton, Reversible jump sampler for autoregressive time series, employing full conditionals to achieve efficient model space moves (1997) |
| [26] | Vermaak, Reversible jump Markov chain Monte Carlo strategies for Bayesian model selection in autoregressive processes, J. Time Ser. Anal. 25 pp 785– (2004) · Zbl 1062.62206 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
