{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T14:18:43Z","timestamp":1781187523242,"version":"3.54.1"},"reference-count":0,"publisher":"Cambridge University Press (CUP)","issue":"4","license":[{"start":{"date-parts":[[2002,5,17]],"date-time":"2002-05-17T00:00:00Z","timestamp":1021593600000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/www.cambridge.org\/core\/terms"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Econom. Theory"],"published-print":{"date-parts":[[2002,8]]},"abstract":"<jats:p>Shibata (1981, <jats:italic>Biometrika<\/jats:italic> 68, 45\u201354) considers \ndata-generating mechanisms belonging to a certain class of linear \nregressions with errors that are independent and identically \nnormally distributed. He compares the variable selection criteria \nAIC (Akaike information criterion) and BIC (Bayesian information \ncriterion) using the following type of comparison. For each \nfixed possible data\u2013generating mechanism, these criteria \nare compared as the data length increases. The results of this \ncomparison have been interpreted as meaning that, in the context \nof the data-generating mechanisms considered by Shibata, AIC \nis better than BIC for large data lengths. Shibata's \ncomparison is pointwise in the space of data\u2013generating \nmechanisms (as the data length increases). Such comparisons \nare potentially misleading. We consider a simple class of \ndata-generating mechanisms satisfying Shibata's assumptions \nand carry out a different type of comparison. For each fixed \ndata length (possibly large) we compare the variable selection \ncriteria for every possible data-generating mechanism in this \nclass. According to this comparison, for this class of \ndata-generating mechanisms no matter how large the data length \nAIC is not better than BIC.<\/jats:p>","DOI":"10.1017\/s0266466602184052","type":"journal-article","created":{"date-parts":[[2002,11,14]],"date-time":"2002-11-14T22:02:47Z","timestamp":1037311367000},"page":"913-925","source":"Crossref","is-referenced-by-count":14,"title":["ON VARIABLE SELECTION IN LINEAR REGRESSION"],"prefix":"10.1017","volume":"18","author":[{"given":"Paul","family":"Kabaila","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"56","published-online":{"date-parts":[[2002,5,17]]},"container-title":["Econometric Theory"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.cambridge.org\/core\/services\/aop-cambridge-core\/content\/view\/S0266466602184052","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2019,4,1]],"date-time":"2019-04-01T19:28:30Z","timestamp":1554146910000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.cambridge.org\/core\/product\/identifier\/S0266466602184052\/type\/journal_article"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2002,5,17]]},"references-count":0,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2002,8]]}},"alternative-id":["S0266466602184052"],"URL":"https:\/\/doi.org\/10.1017\/s0266466602184052","relation":{},"ISSN":["0266-4666","1469-4360"],"issn-type":[{"value":"0266-4666","type":"print"},{"value":"1469-4360","type":"electronic"}],"subject":[],"published":{"date-parts":[[2002,5,17]]}}}