thumbnail

Model selection bias and Freedman's paradox

Annals of the Institute of Statistical Mathematics

By:
, , and
DOI: 10.1007/s10463-009-0234-4

Links

Abstract

In situations where limited knowledge of a system exists and the ratio of data points to variables is small, variable selection methods can often be misleading. Freedman (Am Stat 37:152-155, 1983) demonstrated how common it is to select completely unrelated variables as highly "significant" when the number of data points is similar in magnitude to the number of variables. A new type of model averaging estimator based on model selection with Akaike's AIC is used with linear regression to investigate the problems of likely inclusion of spurious effects and model selection bias, the bias introduced while using the data to select a single seemingly "best" model from a (often large) set of models employing many predictor variables. The new model averaging estimator helps reduce these problems and provides confidence interval coverage at the nominal level while traditional stepwise selection has poor inferential properties. ?? The Institute of Statistical Mathematics, Tokyo 2009.

Additional Publication Details

Publication type:
Article
Publication Subtype:
Journal Article
Title:
Model selection bias and Freedman's paradox
Series title:
Annals of the Institute of Statistical Mathematics
DOI:
10.1007/s10463-009-0234-4
Volume
62
Issue:
1
Year Published:
2010
Language:
English
Larger Work Type:
Article
Larger Work Subtype:
Journal Article
Larger Work Title:
Annals of the Institute of Statistical Mathematics
First page:
117
Last page:
125
Number of Pages:
9