Rights statement: © 2013 International Society for Bayesian Analysis
Final published version, 289 KB, PDF document
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - An adaptive sequential Monte Carlo sampler
AU - Fearnhead, Paul
AU - Taylor, Benjamin M.
N1 - © 2013 International Society for Bayesian Analysis
PY - 2013
Y1 - 2013
N2 - Sequential Monte Carlo (SMC) methods are not only a popular toolin the analysis of state–space models, but offer an alternative to Markov chainMonte Carlo (MCMC) in situations where Bayesian inference must proceed viasimulation. This paper introduces a new SMC method that uses adaptive MCMC kernels for particle dynamics. The proposed algorithm features an online stochastic optimization procedure to select the best MCMC kernel and simultaneously learn optimal tuning parameters. Theoretical results are presented that justify the approach and give guidance on how it should be implemented. Empirical results, based on analysing data from mixture models, show that the new adaptive SMC algorithm (ASMC) can both choose the best MCMC kernel, and learn an appropriate scaling for it. ASMC with a choice between kernels outperformed the adaptive MCMC algorithm of Haario et al. (1998) in 5 out of the 6 cases considered.
AB - Sequential Monte Carlo (SMC) methods are not only a popular toolin the analysis of state–space models, but offer an alternative to Markov chainMonte Carlo (MCMC) in situations where Bayesian inference must proceed viasimulation. This paper introduces a new SMC method that uses adaptive MCMC kernels for particle dynamics. The proposed algorithm features an online stochastic optimization procedure to select the best MCMC kernel and simultaneously learn optimal tuning parameters. Theoretical results are presented that justify the approach and give guidance on how it should be implemented. Empirical results, based on analysing data from mixture models, show that the new adaptive SMC algorithm (ASMC) can both choose the best MCMC kernel, and learn an appropriate scaling for it. ASMC with a choice between kernels outperformed the adaptive MCMC algorithm of Haario et al. (1998) in 5 out of the 6 cases considered.
KW - Adaptive MCMC
KW - Adaptive Sequential Monte Carlo
KW - Bayesian Mixture Analysis
KW - Optimal Scaling
KW - Stochastic Optimization
U2 - 10.1214/13-BA814
DO - 10.1214/13-BA814
M3 - Journal article
VL - 8
SP - 411
EP - 438
JO - Bayesian Analysis
JF - Bayesian Analysis
SN - 1931-6690
IS - 2
ER -