Journal article
Discriminant analysis using the unweighted sum of binary variables: a comparison of model selection methods
Statistics in medicine, Vol.16(23), pp.2679-2700
12/15/1997
DOI: 10.1002/(SICI)1097-0258(19971215)16:23<2679::AID-SIM695>3.0.CO;2-1
PMID: 9421869
Abstract
Many clinical decision‐making rules are equivalent to linear discriminant functions that involve the unweighted sum of binary variables (SBV). We briefly consider the geometry of this restriction and then propose a number of methods for forward stepwise selection of SBV models. Using a simulation study, we compare the performance of these methods under a wide range of plausible conditions and show that no single method is uniformly superior for selecting models of a fixed size. Factors of general importance in relative method performance are the ratio of sample size to the number of candidate variables and the class‐conditional moment structure of the data. We conclude by offering some practical strategies for SBV model construction. © 1997 John Wiley & Sons, Ltd.
Details
- Title: Subtitle
- Discriminant analysis using the unweighted sum of binary variables: a comparison of model selection methods
- Creators
- Douglas R LangbehnRobert F Woolson
- Resource Type
- Journal article
- Publication Details
- Statistics in medicine, Vol.16(23), pp.2679-2700
- DOI
- 10.1002/(SICI)1097-0258(19971215)16:23<2679::AID-SIM695>3.0.CO;2-1
- PMID
- 9421869
- NLM abbreviation
- Stat Med
- ISSN
- 0277-6715
- eISSN
- 1097-0258
- Publisher
- John Wiley & Sons, Ltd; Chichester, UK
- Number of pages
- 22
- Grant note
- United States Public Health Service National Institute of Mental Health (2 T32 MH15168)
- Language
- English
- Date published
- 12/15/1997
- Academic Unit
- Statistics and Actuarial Science; Psychiatry; Epidemiology; Iowa Neuroscience Institute; Biostatistics
- Record Identifier
- 9984004084302771
Metrics
29 Record Views