Active Set and EM Algorithms for Log-Concave Densities Based on Complete and Censored Data

TL;DR

This paper introduces an active set algorithm for estimating log-concave densities from complete data and extends it with an EM algorithm to handle censored or binned data, improving computational efficiency and applicability.

Contribution

The paper presents a novel active set algorithm for log-concave density estimation and integrates it into an EM framework for censored data, advancing existing methods.

Findings

Efficient active set algorithm for complete data

EM algorithm for censored or binned data

Improved computational performance

Abstract

We develop an active set algorithm for the maximum likelihood estimation of a log-concave density based on complete data. Building on this fast algorithm, we indidate an EM algorithm to treat arbitrarily censored or binned data.