On Selecting an Optimal Subset of Regression Variables.
Abstract
In the past decade a number of methods have been developed for selecting the best or at least a good subset of variables in regression analysis. For various reasons, there may be interest in including only a subset say, of size r<p, the number of independent variables. Various authors have considered this problem and a variety of techniques are presently being used to construct such subsets. Most of these seem to lack justification in terms of statistical theory. This paper is interested in deriving a selection procedure to select a random size optimal subset such that all inferior independent variables are excluded. Some results on the efficiency of the procedure are also discussed.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1977
- Accession Number
- ADA044889
Entities
People
- D. Y. Huang
- Shanti Gupta
Organizations
- Purdue University