Maximum likelihood estimation for totally positive log‐concave densities

Abstract

We study nonparametric maximum likelihood estimation for two classes of multivariate distributions that imply strong forms of positive dependence; namely log‐supermodular (MTP2) distributions and log‐L♮‐concave (LLC) distributions. In both cases we also assume log‐concavity in order to ensure boundedness of the likelihood function. Given n independent and identically distributed random vectors from one of our distributions, the maximum likelihood estimator (MLE) exists a.s. and is unique a.e. with probability one when n≥3. This holds independently of the ambient dimension d. We conjecture that the MLE is always the exponential of a tent function. We prove this result for samples in {0,1}d or in under MTP2, and for samples in under LLC. Finally, we provide a conditional gradient algorithm for computing the maximum likelihood estimate.

Document Details

Document Type
Pub Defense Publication
Publication Date
Sep 14, 2020
Source ID
10.1111/sjos.12462

Entities

People

  • Bernd Sturmfels
  • Caroline Uhler
  • Elina Robeva
  • Ngoc Tran

Organizations

  • Alfred P. Sloan Foundation
  • Massachusetts Institute of Technology
  • National Science Foundation Division of Mathematical Sciences
  • Office of Naval Research Global
  • University of British Columbia
  • University of California
  • University of Texas at Austin

Tags

Fields of Study

  • Mathematics

Readers

  • Mathematical Modeling and Probability Theory.
  • Statistical inference.