Asymptotically optimal function estimation by minimum complexity criteria

Andrew Barron, Yuhong Yang, Bin Yu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Scopus citations

Abstract

The minimum description length principle applied to function estimation can yield a criterion of the form - log(likelihood) + const · m instead of the familiar - log(likelihood) + (m/2) log n where m is the number of parameters and n is the sample size. The improved criterion yields minimax optimal rates for redundancy and statistical risk. The analysis suggests an information-theoretic reconciliation of criteria proposed by Rissanen, Schwarz, and Akaike.

Original languageEnglish (US)
Title of host publicationIEEE International Symposium on Information Theory - Proceedings
PublisherIEEE
StatePublished - Dec 1 1994
EventProceedings of the 1994 IEEE International Symposium on Information Theory - Trodheim, Norw
Duration: Jun 27 1994Jul 1 1994

Other

OtherProceedings of the 1994 IEEE International Symposium on Information Theory
CityTrodheim, Norw
Period6/27/947/1/94

    Fingerprint

Cite this

Barron, A., Yang, Y., & Yu, B. (1994). Asymptotically optimal function estimation by minimum complexity criteria. In IEEE International Symposium on Information Theory - Proceedings IEEE.