MODEL SELECTION FOR MULTIVARIATE REGRESSION IN SMALL SAMPLES

Authors
BEDRICK EJ TSAI CL
Citation
Ej. Bedrick et Cl. Tsai, MODEL SELECTION FOR MULTIVARIATE REGRESSION IN SMALL SAMPLES, Biometrics, 50(1), 1994, pp. 226-231
Citations number
13
Categorie Soggetti
Statistic & Probability","Statistic & Probability
Journal title
Biometrics → ACNP
ISSN journal
0006341X
Volume
50
Issue
1
Year of publication
1994
Pages
226 - 231
Database
ISI
SICI code
0006-341X(1994)50:1<226:MSFMRI>2.0.ZU;2-L
Abstract
We develop a small-sample criterion (AIC(C)) for selecting multivariat e regression models. This criterion adjusts the Akaike information cri terion to be an exact unbiased estimator for the expected Kullback-Lei bler information. A small-sample comparison shows that AIC(C) provides better model order choices than other available model selection metho ds. Data from an agricultural experiment are analyzed.