Robustness of likelihood ratio tests: hypothesis testing under incorrect models

Author

Köse, Cenk ; Wesel, Richard D.

Author_Institution

Dept. of Electr. Eng., California Univ., Los Angeles, CA, USA

Volume

2

fYear

2001

fDate

4-7 Nov. 2001

Firstpage

1738

Abstract

This paper formulates the trade-off between the sample size and the reliability of decisions in a multihypothesis test with possible model errors. We introduce a measure of the model uncertainty and derive the best error exponent for the decision error probabilities for the allowed levels of model uncertainty. For small deviations from the correct model, the error exponent for sequential probability ratio tests is quantified.

Keywords

design of experiments; error statistics; signal sampling; Fano´s inequality; Kullback-Leibler information; decision error probability; decision reliability; error exponent; experiment design; heuristic measure; hypothesis testing; incorrect models; likelihood ratio tests; model selection; model uncertainty; model uncertainty measure; multihypothesis test; sample size; sequential probability ratio tests; Bayesian methods; Distortion measurement; Error correction; Error probability; Measurement uncertainty; Minimax techniques; Performance evaluation; Robustness; Sequential analysis; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Signals, Systems and Computers, 2001. Conference Record of the Thirty-Fifth Asilomar Conference on

Conference_Location

Pacific Grove, CA, USA

ISSN

1058-6393

Print_ISBN

0-7803-7147-X

Type

conf

DOI

10.1109/ACSSC.2001.987782

Filename

987782