Efficient estimation of likelihood ratios in state-space models–application to scalable exact approximations of MCMCs

Yıldırım, Sinan and Andrieu, Christophe and Doucet, Arnaud (2015) Efficient estimation of likelihood ratios in state-space models–application to scalable exact approximations of MCMCs. [Working Paper / Technical Report] Sabanci University ID:UNSPECIFIED

[thumbnail of Working paper] PDF (Working paper)
AIS_MCMC_HMM_v1_CA.pdf
Restricted to Repository staff only

Download (1MB) | Request a copy

Abstract

We present a new class of exact-approximate Markov chain Monte Carlo methods for static parameter estimation in hidden Markov models. Like the particle marginal Metropolis-Hastings (PMMH) algorithm, the proposed algorithms target the joint posterior distribution of the latent states while they ‘mimic’ the marginal Metropolis-Hastings algorithm for the static parameter. In contrast to PMMH, these algorithms estimate the acceptance ratio directly without having to use of the same estimate in its denominator, which is a well-known cause for stickiness in pseudo-marginal type algorithms.
Item Type: Working Paper / Technical Report
Subjects: Q Science > QA Mathematics > QA273-280 Probabilities. Mathematical statistics
Divisions: Faculty of Engineering and Natural Sciences > Academic programs > Industrial Engineering
Faculty of Engineering and Natural Sciences
Depositing User: Sinan Yıldırım
Date Deposited: 23 Dec 2015 17:12
Last Modified: 26 Apr 2022 10:53
URI: https://research.sabanciuniv.edu/id/eprint/28575

Actions (login required)

View Item
View Item