Volume 9, June 2005
|Page(s)||220 - 229|
|Published online||15 November 2005|
Risk bounds for mixture density estimation
Center for Biological and Computational Learning, Massachusetts Institute of Technology, Cambridge, MA 02139, USA; firstname.lastname@example.org
2 Department of Mathematics, Massachusetts Institute of Technology, Cambridge, MA 02143, USA.
3 Institute of Statistics and Decision Sciences, Institute for Genome Sciences and Policy, Duke University, Durham, NC 27708, USA.
In this paper we focus on the problem of estimating a bounded density using a finite combination of densities from a given class. We consider the Maximum Likelihood Estimator (MLE) and the greedy procedure described by Li and Barron (1999) under the additional assumption of boundedness of densities. We prove an bound on the estimation error which does not depend on the number of densities in the estimated combination. Under the boundedness assumption, this improves the bound of Li and Barron by removing the factor and also generalizes it to the base classes with converging Dudley integral.
Mathematics Subject Classification: 62G05 / 62G07 / 62G20
Key words: Mixture density estimation / maximum likelihood / Rademacher processes.
© EDP Sciences, SMAI, 2005
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.