File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1093/biomet/asx078
- Scopus: eid_2-s2.0-85048687482
- WOS: WOS:000434111200003
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: On the asymptotic efficiency of approximate Bayesian computation estimators
Title | On the asymptotic efficiency of approximate Bayesian computation estimators |
---|---|
Authors | |
Keywords | Approximate Bayesian computation Proposal distribution Partial information Importance sampling Dimension reduction |
Issue Date | 2018 |
Citation | Biometrika, 2018, v. 105, n. 2, p. 285-299 How to Cite? |
Abstract | © 2018 Biometrika Trust. Many statistical applications involve models for which it is difficult to evaluate the likelihood, but from which it is relatively easy to sample. Approximate Bayesian computation is a likelihood-free method for implementing Bayesian inference in such cases. We present results on the asymptotic variance of estimators obtained using approximate Bayesian computation in a large data limit. Our key assumption is that the data are summarized by a fixed-dimensional summary statistic that obeys a central limit theorem. We prove asymptotic normality of the mean of the approximate Bayesian computation posterior. This result also shows that, in terms of asymptotic variance, we should use a summary statistic that is of the same dimension as the parameter vector, p, and that any summary statistic of higher dimension can be reduced, through a linear transformation, to dimension p in a way that can only reduce the asymptotic variance of the posterior mean. We look at how the Monte Carlo error of an importance sampling algorithm that samples from the approximate Bayesian computation posterior affects the accuracy of estimators. We give conditions on the importance sampling proposal distribution such that the variance of the estimator will be of the same order as that of the maximum likelihood estimator based on the summary statistics used. This suggests an iterative importance sampling algorithm, which we evaluate empirically on a stochastic volatility model. |
Persistent Identifier | http://hdl.handle.net/10722/267095 |
ISSN | 2021 Impact Factor: 3.028 2020 SCImago Journal Rankings: 3.307 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Li, Wentao | - |
dc.contributor.author | Fearnhead, Paul | - |
dc.date.accessioned | 2019-01-31T07:20:30Z | - |
dc.date.available | 2019-01-31T07:20:30Z | - |
dc.date.issued | 2018 | - |
dc.identifier.citation | Biometrika, 2018, v. 105, n. 2, p. 285-299 | - |
dc.identifier.issn | 0006-3444 | - |
dc.identifier.uri | http://hdl.handle.net/10722/267095 | - |
dc.description.abstract | © 2018 Biometrika Trust. Many statistical applications involve models for which it is difficult to evaluate the likelihood, but from which it is relatively easy to sample. Approximate Bayesian computation is a likelihood-free method for implementing Bayesian inference in such cases. We present results on the asymptotic variance of estimators obtained using approximate Bayesian computation in a large data limit. Our key assumption is that the data are summarized by a fixed-dimensional summary statistic that obeys a central limit theorem. We prove asymptotic normality of the mean of the approximate Bayesian computation posterior. This result also shows that, in terms of asymptotic variance, we should use a summary statistic that is of the same dimension as the parameter vector, p, and that any summary statistic of higher dimension can be reduced, through a linear transformation, to dimension p in a way that can only reduce the asymptotic variance of the posterior mean. We look at how the Monte Carlo error of an importance sampling algorithm that samples from the approximate Bayesian computation posterior affects the accuracy of estimators. We give conditions on the importance sampling proposal distribution such that the variance of the estimator will be of the same order as that of the maximum likelihood estimator based on the summary statistics used. This suggests an iterative importance sampling algorithm, which we evaluate empirically on a stochastic volatility model. | - |
dc.language | eng | - |
dc.relation.ispartof | Biometrika | - |
dc.subject | Approximate Bayesian computation | - |
dc.subject | Proposal distribution | - |
dc.subject | Partial information | - |
dc.subject | Importance sampling | - |
dc.subject | Dimension reduction | - |
dc.title | On the asymptotic efficiency of approximate Bayesian computation estimators | - |
dc.type | Article | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1093/biomet/asx078 | - |
dc.identifier.scopus | eid_2-s2.0-85048687482 | - |
dc.identifier.volume | 105 | - |
dc.identifier.issue | 2 | - |
dc.identifier.spage | 285 | - |
dc.identifier.epage | 299 | - |
dc.identifier.eissn | 1464-3510 | - |
dc.identifier.isi | WOS:000434111200003 | - |
dc.identifier.issnl | 0006-3444 | - |