{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T15:46:29Z","timestamp":1785426389545,"version":"3.56.0"},"reference-count":42,"publisher":"Oxford University Press (OUP)","issue":"4","license":[{"start":{"date-parts":[[2020,6,29]],"date-time":"2020-06-29T00:00:00Z","timestamp":1593388800000},"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":[[2020,9,1]]},"abstract":"<jats:title>Summary<\/jats:title><jats:p>Bayesian methods are often optimal, yet increasing pressure for fast computations, especially with streaming data, brings renewed interest in faster, possibly suboptimal, solutions. The extent to which these algorithms approximate Bayesian solutions is a question of interest, but often unanswered. We propose a methodology to address this question in predictive settings, when the algorithm can be reinterpreted as a probabilistic predictive rule. We specifically develop the proposed methodology for a recursive procedure for on-line learning in non-parametric mixture models, which is often referred to as Newton\u2019s algorithm. This algorithm is simple and fast; however, its approximation properties are unclear. By reinterpreting it as a predictive rule, we can show that it underlies a statistical model which is, asymptotically, a Bayesian, exchangeable mixture model. In this sense, the recursive rule provides a quasi-Bayes solution. Although the algorithm offers only a point estimate, our clean statistical formulation enables us to provide the asymptotic posterior distribution and asymptotic credible intervals for the mixing distribution. Moreover, it gives insights for tuning the parameters, as we illustrate in simulation studies, and paves the way to extensions in various directions. Beyond mixture models, our approach can be applied to other predictive algorithms.<\/jats:p>","DOI":"10.1111\/rssb.12385","type":"journal-article","created":{"date-parts":[[2020,6,30]],"date-time":"2020-06-30T06:03:05Z","timestamp":1593496985000},"page":"1087-1114","source":"Crossref","is-referenced-by-count":19,"title":["Quasi-Bayes Properties of a Procedure for Sequential Learning in Mixture Models"],"prefix":"10.1093","volume":"82","author":[{"given":"Sandra","family":"Fortini","sequence":"first","affiliation":[{"name":"Bocconi University , Milan , Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sonia","family":"Petrone","sequence":"additional","affiliation":[{"name":"Bocconi University , Milan , Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2020,6,29]]},"reference":[{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"1466","DOI":"10.1080\/01621459.2014.950735","article-title":"Generalized species sampling priors with latent Beta reinforcements","volume":"109","author":"Airoldi","year":"2014","journal-title":"J. Am. Statist. Ass."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","DOI":"10.1007\/BFb0099421","article-title":"Exchangeability and related topics.","volume-title":"In \u00c9cole d\u2019\u00c9t\u00e9 de Probabilit\u00e9s de Saint-Fleur XIII 1983","author":"Aldous","year":"1985"},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"1152","DOI":"10.1214\/aos\/1176342871","article-title":"Mixtures of Dirichlet processes with applications to Bayesian nonparametric problems","volume":"2","author":"Antoniak","year":"1974","journal-title":"Ann. Statist."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"3243","DOI":"10.3150\/16-BEJ844","article-title":"P\u00f3lya urn schemes with infinitely many colors","volume":"23","author":"Bandyopadhyay","year":"2017","journal-title":"Bernoulli"},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"433","DOI":"10.1239\/aap\/1275055237","article-title":"Conditionally identically distributed species sampling sequences","volume":"42","author":"Bassetti","year":"2010","journal-title":"Adv. Appl. Probab."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"2029","DOI":"10.1214\/009117904000000676","article-title":"Limit theorems for a class of identically distributed random variables","volume":"32","author":"Berti","year":"2004","journal-title":"Ann. Probab."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"2090","DOI":"10.1214\/12-AOP786","article-title":"Exchangeable sequences driven by absolutely continuous random measures","volume":"41","author":"Berti","year":"2013","journal-title":"Ann. Probab."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"353","DOI":"10.1214\/aos\/1176342372","article-title":"Ferguson distributions via P\u00f3lya urn schemes","volume":"1","author":"Blackwell","year":"1973","journal-title":"Ann. Statist."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"859","DOI":"10.1080\/01621459.2017.1285773","article-title":"Variational inference: a review for statisticians","volume":"112","author":"Blei","year":"2017","journal-title":"J. Am. Statist. Ass."},{"key":"2023022408365275700_","first-page":"1727","volume-title":"In Proc. 26th Int. Conf. Neural Information Processing Systems","author":"Broderick","year":"2013"},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"777","DOI":"10.1007\/s11009-017-9587-y","article-title":"A Bayesian motivated Laplace inversion for multivariate probability distributions","volume":"20","author":"Cappello","year":"2018","journal-title":"Methodol. Comp. Appl. Probab."},{"key":"2023022408365275700_","first-page":"1139","article-title":"An almost sure conditional convergence result and an application to a generalized P\u00f3lya urn","volume":"23","author":"Crimaldi","year":"2009","journal-title":"Int. Math. Forum"},{"key":"2023022408365275700_","article-title":"Permutation-based uncertainty quantification about a mixing distribution","author":"Dixit","year":"2019"},{"key":"2023022408365275700_","doi-asserted-by":"crossref","DOI":"10.1002\/9780470316658","volume-title":"Markov Processes: Characterization and Convergence","author":"Ethier","year":"1986"},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"1002","DOI":"10.3150\/11-BEJ356","article-title":"A class of measure-valued Markov chains and Bayesian nonparametrics","volume":"18","author":"Favaro","year":"2012","journal-title":"Bernoulli"},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"423","DOI":"10.1214\/11-BJPS176","article-title":"Predictive construction of priors in Bayesian nonparametrics","volume":"26","author":"Fortini","year":"2012","journal-title":"Braz. J. Probab. Statist."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"819","DOI":"10.1016\/j.spa.2017.06.008","article-title":"On a notion of partially conditionally identically distributed sequences","volume":"128","author":"Fortini","year":"2017","journal-title":"Stoch. Processes. Appl."},{"key":"2023022408365275700_","first-page":"188","article-title":"Minimax multiple shrinkage estimation","volume":"1","author":"George","year":"1986","journal-title":"Ann. Statist."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"429","DOI":"10.1142\/9781860948886_0019","volume-title":"In Frontiers in Statistics","author":"Ghosh","year":"2006"},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"1085","DOI":"10.1080\/01621459.2017.1304219","article-title":"On recursive Bayesian predictive distributions","volume":"113","author":"Hahn","year":"2018","journal-title":"J. Am. Statist. Ass."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-18329-9","volume-title":"Stable Convergence and Stable Limit Theorems","author":"H\u00e4usler","year":"2015"},{"key":"2023022408365275700_","first-page":"1","article-title":"Random replacements in P\u00f3lya urns with infinitely many colours","volume":"24","author":"Janson","year":"2019","journal-title":"Electron. Communs Probab."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"508","DOI":"10.1214\/aop\/1176991771","article-title":"Spreading and predictable sampling in exchangeable sequences and processes","volume":"16","author":"Kallenberg","year":"1988","journal-title":"Ann. Probab."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"3071","DOI":"10.1214\/18-EJS1479","article-title":"A quasi-Bayesian perspective to online clustering","volume":"12","author":"Li","year":"2018","journal-title":"Electron. J. Statist."},{"key":"2023022408365275700_","first-page":"395","volume-title":"In Proc. 26th Int. Conf. Neural Information Processing Systems","author":"Lin","year":"2013"},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"251","DOI":"10.2307\/3315637","article-title":"Importance sampling for nonparametric Bayes models: the next generation","volume":"27","author":"MacEachern","year":"1999","journal-title":"Can. J. Statist."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1214\/17-EJP47","article-title":"Measure-valued P\u00f3lya urn processes","volume":"22","author":"Mailler","year":"2017","journal-title":"Electron. J. Probab."},{"key":"2023022408365275700_","article-title":"A survey of nonparametric mixing density estimation via the predictive recursion algorithm","author":"Martin","year":"2019","journal-title":"Sankhya B"},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"365","DOI":"10.1214\/08-STS265","article-title":"Stochastic approximation and Newtons estimate of a mixing distribution","volume":"23","author":"Martin","year":"2008","journal-title":"Statist. Sci."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"1455","DOI":"10.1214\/09-EJS458","article-title":"Asymptotic properties of predictive recursion: robustness and rate of convergence","volume":"3","author":"Martin","year":"2009","journal-title":"Electron. J. Statist."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"567","DOI":"10.1093\/biomet\/asr030","article-title":"Semiparametric inference in mixture models with predictive recursion marginal likelihood","volume":"98","author":"Martin","year":"2011","journal-title":"Biometrika"},{"key":"2023022408365275700_","first-page":"968","volume-title":"In Proc. 21st Int. Conf. Artificial Intelligence and Statistics","author":"Naesseth","year":"2018"},{"key":"2023022408365275700_","first-page":"306","article-title":"On a nonparametric recursive estimator of the mixing distribution","volume":"64","author":"Newton","year":"2002","journal-title":"Sankhya"},{"key":"2023022408365275700_","volume-title":"In Practical Nonparametric and Semiparametric Bayesian Statistics","author":"Newton","year":"1998"},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1093\/biomet\/86.1.15","article-title":"A recursive algorithm for nonparametric analysis with missing data","volume":"86","author":"Newton","year":"1999","journal-title":"Biometrika"},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/BF02511443","article-title":"Non parametric mixture priors based on an exponential random scheme","volume":"11","author":"Petrone","year":"2002","journal-title":"Statist. Meth. Appl."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"711","DOI":"10.1080\/10618600.2000.10474909","article-title":"Computational aspects of nonparametric Bayesian analysis with applications to the modeling of multiple binary sequences","volume":"9","author":"Quintana","year":"2000","journal-title":"J. Computnl Graph. Statist."},{"key":"2023022408365275700_","first-page":"293","article-title":"On stable sequences of events","volume":"25","author":"Renyi","year":"1963","journal-title":"Sankhya"},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"106","DOI":"10.1111\/j.2517-6161.1978.tb01654.x","article-title":"A quasi-Bayes sequential procedure for mixtures","volume":"40","author":"Smith","year":"1978","journal-title":"J. R. Statist. Soc."},{"key":"2023022408365275700_","first-page":"1391","article-title":"Frequentist coverage of adaptive nonparametric Bayesian credible sets","volume":"43","author":"Szab\u00f3","year":"2015","journal-title":"Ann. Statist."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"2502","DOI":"10.1214\/08-AOS639","article-title":"Consistency of a recursive estimate of mixing distributions","volume":"37","author":"Tokdar","year":"2009","journal-title":"Ann. Statist."},{"key":"2023022408365275700_","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1007\/s11222-018-9803-9","article-title":"Bayesian nonparametric clustering for large data sets","volume":"29","author":"Zuanetti","year":"2019","journal-title":"Statist. Comput."}],"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%2Frssb.12385","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/rssb.12385","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1111\/rssb.12385","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/academic.oup.com\/jrsssb\/article-pdf\/82\/4\/1087\/49323934\/jrsssb_82_4_1087.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jrsssb\/article-pdf\/82\/4\/1087\/49323934\/jrsssb_82_4_1087.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,8]],"date-time":"2024-08-08T19:48:22Z","timestamp":1723146502000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jrsssb\/article\/82\/4\/1087\/7056042"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,6,29]]},"references-count":42,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2020,6,29]]},"published-print":{"date-parts":[[2020,9,1]]}},"URL":"https:\/\/doi.org\/10.1111\/rssb.12385","relation":{},"ISSN":["1369-7412","1467-9868"],"issn-type":[{"value":"1369-7412","type":"print"},{"value":"1467-9868","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2020,9]]},"published":{"date-parts":[[2020,6,29]]}}}