Open Access
October 2006 From ɛ-entropy to KL-entropy: Analysis of minimum information complexity density estimation
Tong Zhang
Ann. Statist. 34(5): 2180-2210 (October 2006). DOI: 10.1214/009053606000000704

Abstract

We consider an extension of ɛ-entropy to a KL-divergence based complexity measure for randomized density estimation methods. Based on this extension, we develop a general information-theoretical inequality that measures the statistical complexity of some deterministic and randomized density estimators. Consequences of the new inequality will be presented. In particular, we show that this technique can lead to improvements of some classical results concerning the convergence of minimum description length and Bayesian posterior distributions. Moreover, we are able to derive clean finite-sample convergence bounds that are not obtainable using previous approaches.

Citation

Download Citation

Tong Zhang. "From ɛ-entropy to KL-entropy: Analysis of minimum information complexity density estimation." Ann. Statist. 34 (5) 2180 - 2210, October 2006. https://doi.org/10.1214/009053606000000704

Information

Published: October 2006
First available in Project Euclid: 23 January 2007

zbMATH: 1106.62005
MathSciNet: MR2291497
Digital Object Identifier: 10.1214/009053606000000704

Subjects:
Primary: 62C10 , 62G07

Keywords: Bayesian posterior distribution , Density estimation , minimum description length

Rights: Copyright © 2006 Institute of Mathematical Statistics

Vol.34 • No. 5 • October 2006
Back to Top