Journal article
A regression model selection criterion based on bootstrap bumping for use with resistant fitting
Computational statistics & data analysis, Vol.35(2), pp.155-169
2000
DOI: 10.1016/S0167-9473(00)00010-4
Abstract
We propose a model selection criterion for regression applications where resistant fitting is appropriate. Our criterion gauges the adequacy of a fitted model based on the median squared error of prediction. The criterion is easily computed using the bootstrap “bumping” algorithm of Tibshirani and Knight (1999, Journal of Computational and Graphical Statistics, pp. 671–686) which provides a convenient method for obtaining least median of squares model parameter estimates. We present an example to illustrate the merit of the criterion in instances where the underlying data set contains influential values. Additionally, we present and discuss the results of a simulation study which illustrates the effectiveness of the criterion under a wide range of error distributions.
Details
- Title: Subtitle
- A regression model selection criterion based on bootstrap bumping for use with resistant fitting
- Creators
- Andrew A Neath - Department of Mathematics and Statistics, P.O. Box 1653, Southern Illinois University, Edwardsville, IL 62026, USAJoseph E Cavanaugh - Department of Statistics, 222 Math Sciences Building, University of Missouri, Columbia, MO 65211, USA
- Resource Type
- Journal article
- Publication Details
- Computational statistics & data analysis, Vol.35(2), pp.155-169
- DOI
- 10.1016/S0167-9473(00)00010-4
- ISSN
- 0167-9473
- eISSN
- 1872-7352
- Publisher
- Elsevier B.V
- Language
- English
- Date published
- 2000
- Academic Unit
- Statistics and Actuarial Science; Biostatistics; Injury Prevention Research Center
- Record Identifier
- 9984214680902771
Metrics
47 Record Views