Remote Access Proceedings of the American Mathematical Society
Green Open Access

Proceedings of the American Mathematical Society

ISSN 1088-6826(online) ISSN 0002-9939(print)



Bias and variance reduction in estimation of model dimension

Authors: Wei-Yin Loh and Xiaodong Zheng
Journal: Proc. Amer. Math. Soc. 122 (1994), 1263-1272
MSC: Primary 62J05; Secondary 62F11
MathSciNet review: 1211583
Full-text PDF

Abstract | References | Similar Articles | Additional Information

Abstract: The problem of estimating the number of regressors to include in a linear regression model is considered. Estimators based on the final prediction error and Akaike's criterion frequently have large positive bias. Shrinkage correction factors and bootstrapping are used to produce new estimators with reduced bias. The asymptotic bias and mean-squared errors of these estimators are derived analytically. Finite-sample estimates are obtained by simulation.

References [Enhancements On Off] (What's this?)

Similar Articles

Retrieve articles in Proceedings of the American Mathematical Society with MSC: 62J05, 62F11

Retrieve articles in all journals with MSC: 62J05, 62F11

Additional Information

Keywords: Akaike criterion, cross-validation, final prediction error, model selection
Article copyright: © Copyright 1994 American Mathematical Society

American Mathematical Society