Most likely maximum entropy for population analysis: A case study in decompression sickness prevention

We estimate the density of a set of biophysical parameters from region censored observations. We propose a new Maximum Entropy (maxent) estimator formulated as finding the most likely constrained maxent density. By using the Ŕnyi entropy of order two instead of the Shannon entropy, we are lead to a...

Full description

Saved in:
Bibliographic Details
Main Authors: Bennani Youssef, Pronzato Luc, Rendas Maria João
Format: Conference Proceeding
Language:English
Subjects:
Decompression sickness
Density
Diving
Entropy
Linear functions
Maximum entropy
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We estimate the density of a set of biophysical parameters from region censored observations. We propose a new Maximum Entropy (maxent) estimator formulated as finding the most likely constrained maxent density. By using the Ŕnyi entropy of order two instead of the Shannon entropy, we are lead to a quadratic optimization problem with linear inequality constraints that has an efficient numerical solution. We compare the proposed estimator to the NPMLE and to the best fitting maxent solutions in real data from hyperbaric diving, showing that the resulting distribution has better generalization performance than NPMLE or maxent alone.
ISSN:0094-243X
1551-7616
DOI:10.1063/1.4906005