{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T07:16:07Z","timestamp":1791530167625,"version":"4.3.4"},"reference-count":32,"publisher":"Oxford University Press (OUP)","issue":"2","license":[{"start":{"date-parts":[[2005,3,9]],"date-time":"2005-03-09T00:00:00Z","timestamp":1110326400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2005,4,1]]},"abstract":"<jats:title>Summary<\/jats:title><jats:p>The expectation\u2013maximization (EM) algorithm is a popular tool for maximizing likelihood functions in the presence of missing data. Unfortunately, EM often requires the evaluation of analytically intractable and high dimensional integrals. The Monte Carlo EM (MCEM) algorithm is the natural extension of EM that employs Monte Carlo methods to estimate the relevant integrals. Typically, a very large Monte Carlo sample size is required to estimate these integrals within an acceptable tolerance when the algorithm is near convergence. Even if this sample size were known at the onset of implementation of MCEM, its use throughout all iterations is wasteful, especially when accurate starting values are not available. We propose a data-driven strategy for controlling Monte Carlo resources in MCEM. The algorithm proposed improves on similar existing methods by recovering EM's ascent (i.e. likelihood increasing) property with high probability, being more robust to the effect of user-defined inputs and handling classical Monte Carlo and Markov chain Monte Carlo methods within a common framework. Because of the first of these properties we refer to the algorithm as \u2018ascent-based MCEM\u2019. We apply ascent-based MCEM to a variety of examples, including one where it is used to accelerate the convergence of deterministic EM dramatically.<\/jats:p>","DOI":"10.1111\/j.1467-9868.2005.00499.x","type":"journal-article","created":{"date-parts":[[2005,3,9]],"date-time":"2005-03-09T09:49:12Z","timestamp":1110361752000},"page":"235-251","source":"Crossref","is-referenced-by-count":92,"title":["Ascent-Based Monte Carlo Expectation\u2013 Maximization"],"prefix":"10.1093","volume":"67","author":[{"given":"Brian S.","family":"Caffo","sequence":"first","affiliation":[{"name":"Johns Hopkins University , Baltimore , USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wolfgang","family":"Jank","sequence":"additional","affiliation":[{"name":"University of Maryland , College Park , USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Galin L.","family":"Jones","sequence":"additional","affiliation":[{"name":"University of Minnesota , Minneapolis , USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2005,3,9]]},"reference":[{"key":"2023040704575016100_","volume-title":"Probability and Measure","author":"Billingsley","year":"1995"},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"265","DOI":"10.1111\/1467-9868.00176","article-title":"Maximizing generalized linear mixed model likelihoods with an automated Monte Carlo EM algorithm","volume":"61","author":"Booth","year":"1999","journal-title":"J. R. Statist. Soc."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"333","DOI":"10.1177\/1471082X0100100407","article-title":"A survey of Monte Carlo algorithms for maximizing the likelihood of a two-stage hierarchical model","volume":"1","author":"Booth","year":"2001","journal-title":"Statist. Modllng"},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"745","DOI":"10.1093\/biomet\/89.4.745","article-title":"Empirical sup rejection sampling","volume":"89","author":"Caffo","year":"2002","journal-title":"Biometrika"},{"key":"2023040704575016100_","first-page":"26","article-title":"geoRglm\u2014a package for generalised linear spatial models","volume":"2","author":"Christensen","year":"2002","journal-title":"R News"},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1111\/j.2517-6161.1977.tb01600.x","article-title":"Maximum likelihood from incomplete data via the EM algorithm","volume":"39","author":"Dempster","year":"1977","journal-title":"J. R. Statist. Soc."},{"key":"2023040704575016100_","first-page":"299","article-title":"Model-based geostatistics","volume":"47","author":"Diggle","year":"1998","journal-title":"Appl. Statist."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1111\/j.2517-6161.1994.tb01976.x","article-title":"On the convergence of Monte Carlo maximum likelihood calculations","volume":"56","author":"Geyer","year":"1994","journal-title":"J. R. Statist. Soc."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"1102","DOI":"10.1198\/016214501753208762","article-title":"A correlated probit model for joint modeling of clustered binary and continuous responses","volume":"96","author":"Gueorguieva","year":"2001","journal-title":"J. Am. Statist. Ass."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"414","DOI":"10.1006\/jmva.1998.1778","article-title":"Geometric ergodicity of Gibbs and block Gibbs samplers for a hierarchical random effects model","volume":"67","author":"Hobert","year":"1998","journal-title":"J. Multiv. Anal."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"731","DOI":"10.1093\/biomet\/89.4.731","article-title":"On the applicability of regenerative simulation in Markov chain Monte Carlo","volume":"89","author":"Hobert","year":"2002","journal-title":"Biometrika"},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"299","DOI":"10.1080\/10618600.1996.10474713","article-title":"R: a language for data analysis and graphics","volume":"5","author":"Ihaka","year":"1996","journal-title":"J. Comput. Graph. Statist."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"214","DOI":"10.1198\/1061860031338","article-title":"Efficiency of Monte Carlo EM and simulated maximum likelihood in two-stage hierarchical models","volume":"12","author":"Jank","year":"2003","journal-title":"J. Comput. Graph. Statist."},{"key":"2023040704575016100_","volume-title":"Output analysis for Markov chain Monte Carlo simulations","author":"Jones","year":"2004"},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"312","DOI":"10.1214\/ss\/1015346315","article-title":"Honest exploration of intractable probability distributions via Markov chain Monte Carlo","volume":"16","author":"Jones","year":"2001","journal-title":"Statist. Sci."},{"key":"2023040704575016100_","volume-title":"The Advanced Theory of Statistics","author":"Kendall","year":"1958"},{"key":"2023040704575016100_","volume-title":"Numerical Analysis for Statisticians","author":"Lange","year":"1999"},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"422","DOI":"10.1198\/106186001317115045","article-title":"Implementations of the Monte Carlo EM algorithm","volume":"10","author":"Levine","year":"2001","journal-title":"J. Comput. Graph. Statist."},{"key":"2023040704575016100_","volume-title":"SAS System for Mixed Models","author":"Littell","year":"1996"},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"226","DOI":"10.1111\/j.2517-6161.1982.tb01203.x","article-title":"Finding the observed information matrix when using the EM algorithm","volume":"44","author":"Louis","year":"1982","journal-title":"J. R. Statist. Soc."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"330","DOI":"10.1080\/01621459.1994.10476474","article-title":"Maximum likelihood variance components estimation for binary data","volume":"89","author":"McCulloch","year":"1994","journal-title":"J. Am. Statist. Ass."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"162","DOI":"10.1080\/01621459.1997.10473613","article-title":"Maximum likelihood algorithms for generalized linear mixed models","volume":"92","author":"McCulloch","year":"1997","journal-title":"J. Am. Statist. Ass."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4471-3267-7","volume-title":"Markov Chains and Stochastic Stability","author":"Meyn","year":"1993"},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"233","DOI":"10.1080\/01621459.1995.10476507","article-title":"Regeneration in Markov chain samplers","volume":"90","author":"Mykland","year":"1995","journal-title":"J. Am. Statist. Ass."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511526237","volume-title":"General Irreducible Markov Chains and Non-negative Operators","author":"Nummelin","year":"1984"},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"838","DOI":"10.1137\/0330046","article-title":"Acceleration of stochastic-approximation by averaging","volume":"30","author":"Polyak","year":"1992","journal-title":"SIAM J. Control Optimzn"},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"429","DOI":"10.1016\/S0167-9473(98)00075-9","article-title":"Monte Carlo EM with importance reweighting and its applications in random effects models","volume":"29","author":"Quintana","year":"1999","journal-title":"Comput. Statist. Data Anal."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4757-3071-5","volume-title":"Monte Carlo Statistical Methods","author":"Robert","year":"1999"},{"key":"2023040704575016100_","doi-asserted-by":"crossref","DOI":"10.1002\/9780470316856","volume-title":"Variance Components","author":"Searle","year":"1992"},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"221","DOI":"10.1111\/1467-9868.00334","article-title":"Publication bias and meta-analysis for 2\u00d72 tables: an average Markov chain Monte Carlo EM algorithm","volume":"64","author":"Shi","year":"2002","journal-title":"J. R. Statist. Soc."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"699","DOI":"10.1080\/01621459.1990.10474930","article-title":"A Monte Carlo implementation of the EM algorithm and the poor man's data augmentation algorithms","volume":"85","author":"Wei","year":"1990","journal-title":"J. Am. Statist. Ass."},{"key":"2023040704575016100_","doi-asserted-by":"crossref","first-page":"95","DOI":"10.1214\/aos\/1176346060","article-title":"On the convergence properties of the EM algorithm","volume":"11","author":"Wu","year":"1983","journal-title":"Ann. Statist."}],"container-title":["Journal of the Royal Statistical Society Series B: Statistical Methodology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1111%2Fj.1467-9868.2005.00499.x","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/academic.oup.com\/jrsssb\/article-pdf\/67\/2\/235\/49795084\/jrsssb_67_2_235.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jrsssb\/article-pdf\/67\/2\/235\/49795084\/jrsssb_67_2_235.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,12,26]],"date-time":"2024-12-26T12:01:47Z","timestamp":1735214507000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jrsssb\/article\/67\/2\/235\/7109470"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2005,3,9]]},"references-count":32,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2005,3,9]]},"published-print":{"date-parts":[[2005,4,1]]}},"URL":"https:\/\/doi.org\/10.1111\/j.1467-9868.2005.00499.x","relation":{},"ISSN":["1369-7412","1467-9868"],"issn-type":[{"value":"1369-7412","type":"print"},{"value":"1467-9868","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2005,4]]},"published":{"date-parts":[[2005,3,9]]}}}