Artigo
A note on model uncertainty in linear regression
Registro en:
Journal Of The Royal Statistical Society Series D-the Statistician. Oxford: Blackwell Publ Ltd, v. 52, p. 165-177, 2003.
0039-0526
10.1111/1467-9884.00349
WOS:000183546800003
Autor
Candolo, C.
Davison, A. C.
Demetrio, CGB
Resumen
We consider model selection uncertainty in linear regression. We study theoretically and by simulation the approach of Buckland and co-workers, who proposed estimating a parameter common to all models under study by taking a weighted average over the models, using weights obtained from information criteria or the bootstrap. This approach is compared with the usual approach in which the 'best' model is used, and with Bayesian model averaging. The weighted predictor behaves similarly to model averaging, with generally more realistic mean-squared errors than the usual model-selection-based estimator. Swiss Fed Inst Technol, Math Inst, CH-1015 Lausanne, Switzerland Univ Fed Sao Carlos, BR-13560 Sao Carlos, SP, Brazil State Univ Sao Paulo, Piracicaba, Brazil State Univ Sao Paulo, Piracicaba, Brazil