A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
Importance sampling type estimators based on approximate marginal Markov chain Monte Carlo (2020)

Vihola, M., Helske, J., & Franks, J. (2020). Importance sampling type estimators based on approximate marginal Markov chain Monte Carlo. Scandinavian Journal of Statistics, 47(4), 1339-1376. https://doi.org/10.1111/sjos.12492

JYU-tekijät tai -toimittajat

Julkaisun tiedot

Julkaisun kaikki tekijät tai toimittajat: Vihola, Matti; Helske, Jouni; Franks, Jordan

Lehti tai sarja: Scandinavian Journal of Statistics

ISSN: 0303-6898

eISSN: 1467-9469

Julkaisuvuosi: 2020

Ilmestymispäivä: 03.09.2020

Volyymi: 47

Lehden numero: 4

Artikkelin sivunumerot: 1339-1376

Kustantaja: Wiley-Blackwell

Julkaisumaa: Britannia

Julkaisun kieli: englanti

DOI: https://doi.org/10.1111/sjos.12492

Julkaisun avoin saatavuus: Ei avoin

Julkaisukanavan avoin saatavuus:

Julkaisu on rinnakkaistallennettu (JYX): https://jyx.jyu.fi/handle/123456789/71682

Julkaisu on rinnakkaistallennettu: https://arxiv.org/abs/1609.02541


We consider importance sampling (IS) type weighted estimators based on Markov chain Monte Carlo (MCMC) targeting an approximate marginal of the target distribution. In the context of Bayesian latent variable models, the MCMC typically operates on the hyperparameters, and the subsequent weighting may be based on IS or sequential Monte Carlo (SMC), but allows for multilevel techniques as well. The IS approach provides a natural alternative to delayed acceptance (DA) pseudo-marginal/particle MCMC, and has many advantages over DA, including a straightforward parallelisation and additional flexibility in MCMC implementation. We detail minimal conditions which ensure strong consistency of the suggested estimators, and provide central limit theorems with expressions for asymptotic variances. We demonstrate how our method can make use of SMC in the state space models context, using Laplace approximations and time-discretised diffusions. Our experimental results are promising and show that the IS type approach can provide substantial gains relative to an analogous DA scheme, and is often competitive even without parallelisation.

YSO-asiasanat: tilastomenetelmät; bayesilainen menetelmä; otanta; estimointi; Markovin ketjut; Monte Carlo -menetelmät

Vapaat asiasanat: Markov chain Monte Carlo (MCMC); Bayesian analysis

Liittyvät organisaatiot

Hankkeet, joissa julkaisu on tehty

OKM-raportointi: Kyllä

Raportointivuosi: 2020

JUFO-taso: 2

Viimeisin päivitys 2022-17-06 klo 12:14