{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T09:33:02Z","timestamp":1790155982210,"version":"4.0.1"},"reference-count":45,"publisher":"Oxford University Press (OUP)","issue":"2","license":[{"start":{"date-parts":[[2003,4,25]],"date-time":"2003-04-25T00:00:00Z","timestamp":1051228800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"funder":[{"DOI":"10.13039\/501100000780","name":"European Union","doi-asserted-by":"publisher","award":["ERB-FMRX-CT96-0095"],"award-info":[{"award-number":["ERB-FMRX-CT96-0095"]}],"id":[{"id":"10.13039\/501100000780","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000266","name":"Engineering and Physical Sciences Research Council","doi-asserted-by":"publisher","award":["AF\/000537"],"award-info":[{"award-number":["AF\/000537"]}],"id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000266","name":"Engineering and Physical Sciences Research Council","doi-asserted-by":"publisher","award":["RG\/M76157"],"award-info":[{"award-number":["RG\/M76157"]}],"id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2003,5,1]]},"abstract":"<jats:title>Summary<\/jats:title><jats:p>The classical approach to statistical analysis is usually based upon finding values for model parameters that maximize the likelihood function. Model choice in this context is often also based on the likelihood function, but with the addition of a penalty term for the number of parameters. Though models may be compared pairwise by using likelihood ratio tests for example, various criteria such as the Akaike information criterion have been proposed as alternatives when multiple models need to be compared. In practical terms, the classical approach to model selection usually involves maximizing the likelihood function associated with each competing model and then calculating the corresponding criteria value(s). However, when large numbers of models are possible, this quickly becomes infeasible unless a method that simultaneously maximizes over both parameter and model space is available. We propose an extension to the traditional simulated annealing algorithm that allows for moves that not only change parameter values but also move between competing models. This transdimensional simulated annealing algorithm can therefore be used to locate models and parameters that minimize criteria such as the Akaike information criterion, but within a single algorithm, removing the need for large numbers of simulations to be run. We discuss the implementation of the transdimensional simulated annealing algorithm and use simulation studies to examine its performance in realistically complex modelling situations. We illustrate our ideas with a pedagogic example based on the analysis of an autoregressive time series and two more detailed examples: one on variable selection for logistic regression and the other on model selection for the analysis of integrated recapture\u2013recovery data.<\/jats:p>","DOI":"10.1111\/1467-9868.00399","type":"journal-article","created":{"date-parts":[[2004,12,28]],"date-time":"2004-12-28T19:38:40Z","timestamp":1104262720000},"page":"503-520","source":"Crossref","is-referenced-by-count":46,"title":["Classical Model Selection via Simulated Annealing"],"prefix":"10.1093","volume":"65","author":[{"given":"S. P.","family":"Brooks","sequence":"first","affiliation":[{"name":"University of Cambridge , UK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"N.","family":"Friel","sequence":"additional","affiliation":[{"name":"University of Glasgow , UK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"R.","family":"King","sequence":"additional","affiliation":[{"name":"University of Cambridge , UK"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2003,4,25]]},"reference":[{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"716","DOI":"10.1109\/TAC.1974.1100705","article-title":"A new look at the statistical identification model","volume":"19","author":"Akaike","year":"1974","journal-title":"IEEE Trans. Autom. Control"},{"key":"2023032908214869900_","article-title":"Reversible jump MCMC simulated annealing for neural networks","author":"Andrieu","year":"2001","journal-title":"Neur. Comput."},{"key":"2023032908214869900_","first-page":"491","volume-title":"Analysis of Statistical Information","author":"Besag","year":"1989"},{"key":"2023032908214869900_","volume-title":"Time Series Analysis: Forecasting and Control","author":"Box","year":"1970"},{"key":"2023032908214869900_","volume-title":"Algorithms for Minimizing without Derivatives","author":"Brent","year":"1973"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"69","DOI":"10.1111\/1467-9884.00117","article-title":"Markov chain Monte Carlo method and its application","volume":"47","author":"Brooks","year":"1998","journal-title":"Statistician"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1111\/1467-9868.03711","article-title":"Efficient construction of reversible jump proposal distributions (with discussion)","volume":"65","author":"Brooks","year":"2003","journal-title":"J. R. Statist. Soc. B"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1007\/BF00142569","article-title":"Automatic starting point selection for function optimisation","volume":"4","author":"Brooks","year":"1994","journal-title":"Statist. Comput."},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"241","DOI":"10.2307\/2348448","article-title":"Optimization using simulated annealing","volume":"44","author":"Brooks","year":"1995","journal-title":"Statistician"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"1097","DOI":"10.2307\/2533567","article-title":"Finite mixture models for proportions","volume":"53","author":"Brooks","year":"1997","journal-title":"Biometrics"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"33","DOI":"10.2307\/2533993","article-title":"Integrated recovery\/recapture data analysis","volume":"54","author":"Catchpole","year":"1998","journal-title":"Biometrics"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"262","DOI":"10.1145\/29380.29864","article-title":"Minimizing multimodal functions of continuous variables with the simulated annealing algorithm","volume":"13","author":"Corana","year":"1987","journal-title":"ACM Trans. Math. Softwr."},{"key":"2023032908214869900_","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 (with discussion)","volume":"39","author":"Dempster","year":"1977","journal-title":"J. R. Statist. Soc. B"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"1541","DOI":"10.1109\/20.123991","article-title":"A combined strategy for optimization in non linear magnetic problems using simulated annealing and search techniques","volume":"28","author":"Drago","year":"1992","journal-title":"IEEE Trans. Magn."},{"key":"2023032908214869900_","article-title":"A case study of stochastic optimization in health policy: problem formulation and preliminary results","author":"Draper","year":"2001","journal-title":"J. Glob. Optimizn"},{"key":"2023032908214869900_","article-title":"Model uncertainty in integrated ARMA processes","volume-title":"Technical Report","author":"Ehlers","year":"2003"},{"key":"2023032908214869900_","volume-title":"Practical Methods of Optimization","author":"Fletcher","year":"1987"},{"key":"2023032908214869900_","volume-title":"Monte Carlo Markov Chains for dynamic generalized linear models","author":"Gamerman","year":"1995"},{"key":"2023032908214869900_","first-page":"145","article-title":"Model determination using sampling-based methods","volume-title":"Markov Chain Monte Carlo in Practice","author":"Gelfand","year":"1996"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"398","DOI":"10.1080\/01621459.1990.10476213","article-title":"Sampling based approaches to calculating marginal densities","volume":"85","author":"Gelfand","year":"1990","journal-title":"J. Am. Statist. Ass."},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"721","DOI":"10.1109\/TPAMI.1984.4767596","article-title":"Stochastic relaxation, Gibbs distributions and the Bayesian restoration of images","volume":"6","author":"Geman","year":"1984","journal-title":"IEEE Trans. Pattn Anal. Mach. Intell."},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"731","DOI":"10.1093\/biomet\/87.4.731","article-title":"Calibration and empirical Bayes variable selection","volume":"87","author":"George","year":"2000","journal-title":"Biometrika"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"73","DOI":"10.1007\/BF01007975","article-title":"Nonstationary Markov chains and the convergence of the annealing algorithm","volume":"39","author":"Gidas","year":"1985","journal-title":"J. Statist. Phys."},{"key":"2023032908214869900_","volume-title":"Markov Chain Monte Carlo in Practice","author":"Gilks","year":"1996"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4615-6089-0","volume-title":"TABU Search","author":"Glover","year":"1997"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"711","DOI":"10.1093\/biomet\/82.4.711","article-title":"Reversible jump Markov chain Monte Carlo computation and Bayesian model determination","volume":"82","author":"Green","year":"1995","journal-title":"Biometrika"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"87","DOI":"10.1016\/0895-7177(92)90108-W","article-title":"Genetic algorithms and very fast simulated re-annealing: a comparison","volume":"16","author":"Ingber","year":"1992","journal-title":"Math. Comput. Modllng"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","DOI":"10.1111\/j.0006-341X.2002.00841.x","article-title":"Model selection for integrated recovery\/recapture data","author":"King","year":"2002","journal-title":"Biometrics"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"975","DOI":"10.1007\/BF01009452","article-title":"Optimization by simulated annealing: quantitative studies","volume":"34","author":"Kirkpatrick","year":"1984","journal-title":"J. Statist. Phys."},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"671","DOI":"10.1126\/science.220.4598.671","article-title":"Optimisation using simulated annealing","volume":"220","author":"Kirkpatrick","year":"1983","journal-title":"Science"},{"key":"2023032908214869900_","first-page":"2311","volume-title":"Encyclopedia of Biostatistics","author":"Lemeshow","year":"1998"},{"key":"2023032908214869900_","first-page":"243","volume-title":"Bayesian Statistics and Econometrics: Essays in Honor of Arnold Zellner","author":"Marriott","year":"1996"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"1087","DOI":"10.1063\/1.1699114","article-title":"Equations of state calculations by fast computing machines","volume":"21","author":"Metropolis","year":"1953","journal-title":"J. Chem. Phys."},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"308","DOI":"10.1093\/comjnl\/7.4.308","article-title":"A simplex method for function minimization","volume":"7","author":"Nelder","year":"1965","journal-title":"Comput. J."},{"key":"2023032908214869900_","first-page":"216","article-title":"Investigation of phase transitions by a Monte Carlo method","volume":"7","author":"Norman","year":"1969","journal-title":"High Temp."},{"key":"2023032908214869900_","volume-title":"Spectral Analysis and Times Series","author":"Priestley","year":"1981"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","DOI":"10.1002\/9780470316726","volume-title":"Stochastic Simulation","author":"Ripley","year":"1987"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"291","DOI":"10.1111\/1467-9868.00070","article-title":"Updating schemes, correlation structure, blocking and parameterization for the Gibbs sampler","volume":"59","author":"Roberts","year":"1997","journal-title":"J. R. Statist. Soc. B"},{"key":"2023032908214869900_","first-page":"427","article-title":"A review of estimating animal abundance III","volume":"14","author":"Schwarz","year":"1998","journal-title":"Statist. Sci."},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"461","DOI":"10.1214\/aos\/1176344136","article-title":"Estimating the dimension of a model","volume":"6","author":"Schwarz","year":"1978","journal-title":"Ann. Statist."},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"1545","DOI":"10.1109\/20.123992","article-title":"Optimizing electromagnetic devices combining direct search methods with simulated annealing","volume":"28","author":"Simkin","year":"1992","journal-title":"IEEE Trans. Magn."},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"201","DOI":"10.1007\/BF00140865","article-title":"Partitioned algorithms for maximum likelihood and other nonlinear estimation","volume":"6","author":"Smyth","year":"1996","journal-title":"Statist. Comput."},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"40","DOI":"10.1214\/aos\/1016120364","article-title":"Bayesian analysis of mixture models with an unknown number of components: an alternative to reversible jump methods","volume":"28","author":"Stephens","year":"2000","journal-title":"Ann. Statist."},{"key":"2023032908214869900_","volume-title":"Reversible jump samplers for autoregressive time series, employing full conditionals to achieve efficient model space moves","author":"Troughton","year":"1997"},{"key":"2023032908214869900_","doi-asserted-by":"crossref","first-page":"259","DOI":"10.1016\/0021-9991(84)90095-0","article-title":"A Monte Carlo simulated annealing approach to optimization over continuous variables","volume":"56","author":"Vanderbilt","year":"1984","journal-title":"J. Comput. Phys."}],"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%2F1467-9868.00399","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1111%2F1467-9868.00399","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/academic.oup.com\/jrsssb\/article-pdf\/65\/2\/503\/49684324\/jrsssb_65_2_503.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jrsssb\/article-pdf\/65\/2\/503\/49684324\/jrsssb_65_2_503.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,12,21]],"date-time":"2024-12-21T21:14:08Z","timestamp":1734815648000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jrsssb\/article\/65\/2\/503\/7092839"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2003,4,25]]},"references-count":45,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2003,4,25]]},"published-print":{"date-parts":[[2003,5,1]]}},"URL":"https:\/\/doi.org\/10.1111\/1467-9868.00399","relation":{},"ISSN":["1369-7412","1467-9868"],"issn-type":[{"value":"1369-7412","type":"print"},{"value":"1467-9868","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2003,5]]},"published":{"date-parts":[[2003,4,25]]}}}