Katalog GES

Speeding up MCMC by delayed acceptance and data subsampling

1st Person: Quiroz, Matias
Additional Persons: Villani, Mattias; Kohn, Robert
Type of Publication: Paper
Language: English
Published: Sveriges Riksbank 2015
Series: Sveriges Riksbank Working Paper Series
Online: https://www.econstor.eu/bitstream/10419/129721/1/832334022.pdf
Description: We propose a generic Markov Chain Monte Carlo (MCMC) algorithm to speed up computations for datasets with many observations. A key feature of our approach is the use of the highly efficient difference estimator from the survey sampling literature to estimate the log-likelihood accurately using only a small fraction of the data. Our algorithm improves on the O(n) complexity of regular MCMC by operating over local data clusters instead of the full sample when computing the likelihood. The likelihood estimate is used in a Pseudo- marginal framework to sample from a perturbed posterior which is within O(m-1/2) of the true posterior, where m is the subsample size. The method is applied to a logistic regression model to predict firm bankruptcy for a large data set. We document a significant speed up in comparison to the standard MCMC on the full dataset.

Similar Items

Cannot find similar records

Library Services

Search Options

Quick links