Approximation by Log-Concave Distributions with Applications to Regression
Authors:
- Lutz DuembgenLDLutz Duembgen
Preprint
Abstract
We study the approximation of arbitrary distributions P on d-dimensional space by distributions with log-concave density. Approximation means minimizing a Kullback-Leibler type functional. We show that such an approximation exists if, and only if, P has finite first moments and is not supported by some hyperplane. Furthermore we show that this approximation depends continuously on P with respect to Mallows' distance D_1. This result implies consistency of the maximum likelihood estimator of a log-concave density under fairly general conditions. It also allows us to prove existence and consistency of estimators in regression models with a response Y = m(X) + E, where X and E are independent, m(.) belongs to a certain class of regression functions while E is a random error with log-concave density.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- Nonparametric estimation of multivariate convex-transformed densitiesThe Annals of Statistics, 2010
- Consistency of multivariate log-concave density estimatorsStatistics & Probability Letters, 2010
- Theoretical properties of the log-concave maximum likelihood estimator of a multidimensional densityElectronic Journal of Statistics, 2010
- Inference and Modeling with Log-concave DistributionsStatistical Science, 2009
- Limit distribution theory for maximum likelihood estimation of a log-concave densityThe Annals of Statistics, 2009
- Thinking outside the box: Statistical inference based on Kullback–Leibler empirical projectionsStatistics & Probability Letters, 2007
- A note on the measurability of convex setsArchiv der Mathematik, 1986
- A Note on Asymptotic Joint NormalityThe Annals of Mathematical Statistics, 1972
- A Simplex Method for Function MinimizationThe Computer Journal, 1965
- An Empirical Distribution Function for Sampling with Incomplete InformationThe Annals of Mathematical Statistics, 1955