{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,11]],"date-time":"2026-10-11T22:15:09Z","timestamp":1791756909763,"version":"4.3.4"},"reference-count":49,"publisher":"Wiley","issue":"7","license":[{"start":{"date-parts":[[2015,10,29]],"date-time":"2015-10-29T00:00:00Z","timestamp":1446076800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2015,9,1]],"date-time":"2015-09-01T00:00:00Z","timestamp":1441065600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"name":"MRC","award":["MR\/J013692\/1"],"award-info":[{"award-number":["MR\/J013692\/1"]}]},{"name":"MRC","award":["U1052 60558"],"award-info":[{"award-number":["U1052 60558"]}]}],"content-domain":{"domain":["wiley.com"],"crossmark-restriction":true},"short-container-title":["Statistics in Medicine"],"published-print":{"date-parts":[[2016,3,30]]},"abstract":"<jats:p>Risk prediction models are used to predict a clinical outcome for patients using a set of predictors. We focus on predicting low\u2010dimensional binary outcomes typically arising in epidemiology, health services and public health research where logistic regression is commonly used. When the number of events is small compared with the number of regression coefficients, model overfitting can be a serious problem. An overfitted model tends to demonstrate poor predictive accuracy when applied to new data. We review frequentist and Bayesian shrinkage methods that may alleviate overfitting by shrinking the regression coefficients towards zero (some methods can also provide more parsimonious models by omitting some predictors). We evaluated their predictive performance in comparison with maximum likelihood estimation using real and simulated data. The simulation study showed that maximum likelihood estimation tends to produce overfitted models with poor predictive performance in scenarios with few events, and penalised methods can offer improvement. Ridge regression performed well, except in scenarios with many noise predictors. Lasso performed better than ridge in scenarios with many noise predictors and worse in the presence of correlated predictors. Elastic net, a hybrid of the two, performed well in all scenarios. Adaptive lasso and smoothly clipped absolute deviation performed best in scenarios with many noise predictors; in other scenarios, their performance was inferior to that of ridge and lasso. Bayesian approaches performed well when the hyperparameters for the priors were chosen carefully. Their use may aid variable selection, and they can be easily extended to clustered\u2010data settings and to incorporate external information. \u00a9 2015 The Authors. Statistics in Medicine Published by JohnWiley &amp; Sons Ltd.<\/jats:p>","DOI":"10.1002\/sim.6782","type":"journal-article","created":{"date-parts":[[2015,10,29]],"date-time":"2015-10-29T23:08:42Z","timestamp":1446160122000},"page":"1159-1177","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":361,"title":["Review and evaluation of penalised regression methods for risk prediction in low\u2010dimensional data with few events"],"prefix":"10.1002","volume":"35","author":[{"given":"Menelaos","family":"Pavlou","sequence":"first","affiliation":[{"name":"Department of Statistical Science University College London London WC1E 6BT U.K."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gareth","family":"Ambler","sequence":"additional","affiliation":[{"name":"Department of Statistical Science University College London London WC1E 6BT U.K."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shaun","family":"Seaman","sequence":"additional","affiliation":[{"name":"MRC Biostatistics Unit Cambridge CB2 0SR U.K."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maria","family":"De\u00a0Iorio","sequence":"additional","affiliation":[{"name":"Department of Statistical Science University College London London WC1E 6BT U.K."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rumana Z","family":"Omar","sequence":"additional","affiliation":[{"name":"Department of Statistical Science University College London London WC1E 6BT U.K."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2015,10,29]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.b375"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.b606"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1136\/heartjnl-2011-301246"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.39609.449676.25"},{"issue":"1","key":"e_1_2_9_6_1","first-page":"743","article-title":"General cardiovascular risk profile for use in primary care: the Framingham Heart Study","volume":"117","author":"D'Agostino RB","journal-title":"Circulation"},{"issue":"1","key":"e_1_2_9_7_1","first-page":"364","article-title":"A novel clinical risk prediction model for sudden cardiac death in hypertrophic cardiomyopathy","volume":"33","author":"O' Mahony C","year":"2013","journal-title":"European Heart Journal"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/S1569-9056(09)60974-X"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4757-3462-1"},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0895-4356(96)00236-3"},{"key":"e_1_2_9_11_1","doi-asserted-by":"publisher","DOI":"10.1093\/aje\/kwk052"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pmed.1001221"},{"key":"e_1_2_9_13_1","doi-asserted-by":"publisher","DOI":"10.1002\/(SICI)1097-0258(20000430)19:8<1059::AID-SIM412>3.0.CO;2-0"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1002\/sim.4371"},{"key":"e_1_2_9_15_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11222-009-9155-6"},{"key":"e_1_2_9_16_1","unstructured":"SortariS.Penalized regression: bootstrap confidence intervals and variable selection for high\u2010dimensional data sets.Ph.D. Thesis Universita degli Studi di Milano 2011."},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.1214\/10-BA607"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1111\/rssb.12080"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1002\/sim.4439"},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1002\/sim.4780122407"},{"key":"e_1_2_9_21_1","doi-asserted-by":"publisher","DOI":"10.2307\/2347628"},{"issue":"4","key":"e_1_2_9_22_1","first-page":"708","article-title":"Ridge regression \u2013 when biased estimation is better","volume":"58","author":"Feig DG","year":"1978","journal-title":"Social Science Quarterly"},{"key":"e_1_2_9_23_1","doi-asserted-by":"crossref","first-page":"267","DOI":"10.1111\/j.2517-6161.1996.tb02080.x","article-title":"Regression shrinkage and selection via the lasso","volume":"58","author":"Tibshirani Robert","year":"1996","journal-title":"Journal of the Royal Statistical Society, Series B"},{"key":"e_1_2_9_24_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1467-9868.2005.00503.x"},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-20192-9"},{"key":"e_1_2_9_26_1","doi-asserted-by":"publisher","DOI":"10.1198\/016214506000000735"},{"key":"e_1_2_9_27_1","doi-asserted-by":"publisher","DOI":"10.1002\/bimj.200900064"},{"key":"e_1_2_9_28_1","doi-asserted-by":"publisher","DOI":"10.1214\/10-AOAS388"},{"key":"e_1_2_9_29_1","doi-asserted-by":"publisher","DOI":"10.1198\/016214501753382273"},{"key":"e_1_2_9_30_1","doi-asserted-by":"publisher","DOI":"10.1002\/sim.3107"},{"key":"e_1_2_9_31_1","unstructured":"GoemanJJ.L1 and l2 penalized regression models Technical Document 2012."},{"key":"e_1_2_9_32_1","doi-asserted-by":"publisher","DOI":"10.1198\/jasa.2011.tm10159"},{"key":"e_1_2_9_33_1","doi-asserted-by":"publisher","DOI":"10.2307\/2684406"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1198\/016214508000000337"},{"key":"e_1_2_9_35_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11222-009-9158-3"},{"key":"e_1_2_9_36_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1467-9876.2010.00723.x"},{"key":"e_1_2_9_37_1","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.1993.10476353"},{"key":"e_1_2_9_38_1","doi-asserted-by":"publisher","DOI":"10.1214\/009053604000001147"},{"key":"e_1_2_9_39_1","doi-asserted-by":"publisher","DOI":"10.1214\/09-BA403"},{"key":"e_1_2_9_40_1","doi-asserted-by":"publisher","DOI":"10.1002\/sim.1422"},{"key":"e_1_2_9_41_1","doi-asserted-by":"publisher","DOI":"10.1097\/EDE.0b013e3181c30fb2"},{"key":"e_1_2_9_42_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-0-387-77244-8"},{"key":"e_1_2_9_43_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.csda.2013.09.008"},{"issue":"4","key":"e_1_2_9_44_1","first-page":"1603","article-title":"Adaptive lasso for sparse high\u2010dimensional regression models","volume":"18","author":"Huang J","year":"2008","journal-title":"Statistica Sinica"},{"key":"e_1_2_9_45_1","doi-asserted-by":"publisher","DOI":"10.1002\/sam.11183"},{"key":"e_1_2_9_46_1","doi-asserted-by":"publisher","DOI":"10.1093\/biomet\/80.1.27"},{"key":"e_1_2_9_47_1","doi-asserted-by":"publisher","DOI":"10.1002\/sim.1047"},{"key":"e_1_2_9_48_1","doi-asserted-by":"publisher","DOI":"10.1186\/1471-2288-14-137"},{"key":"e_1_2_9_49_1","doi-asserted-by":"publisher","DOI":"10.7326\/M14-0697"},{"key":"e_1_2_9_50_1","doi-asserted-by":"publisher","DOI":"10.1613\/jair.953"}],"container-title":["Statistics in Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1002%2Fsim.6782","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/sim.6782","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/sim.6782","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/sim.6782","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,6,11]],"date-time":"2024-06-11T23:42:32Z","timestamp":1718149352000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/sim.6782"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2015,10,29]]},"references-count":49,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2016,3,30]]}},"alternative-id":["10.1002\/sim.6782"],"URL":"https:\/\/doi.org\/10.1002\/sim.6782","archive":["Portico"],"relation":{},"ISSN":["0277-6715","1097-0258"],"issn-type":[{"value":"0277-6715","type":"print"},{"value":"1097-0258","type":"electronic"}],"subject":[],"published":{"date-parts":[[2015,10,29]]}}}