Cross-validation and multinomial prediction

1974404 citationsJournal Article

Authors

Abstract

The method of cross-validatory choice and assessment is applied to prediction of a multinomial indicator. The resulting predictor is compared with analogous expressions due to Good and Fienberg & Holland. Some numerical comparisons are made.

Topics & Keywords

Multi-Criteria Decision Making Statistical and Computational Modeling

Publication Details

Published in: Biometrika

Volume 61, Issue 3, pp. 509-515

DOI: 10.1093/biomet/61.3.509

Field-Weighted Citation Impact: 4.22

Command Palette

Cross-validation and multinomial prediction

Authors

Abstract

Topics & Keywords

Publication Details