Entropy Evaluation Based on Value Validity

Besides its importance in statistical physics and information theory, the Boltzmann-Shannon entropy S has become one of the most widely used and misused summary measures of various attributes (characteristics) in diverse fields of study. It has also been the subject of extensive and perhaps excessiv...

Full description

Bibliographic Details
Main Author: Tarald O. Kvålseth
Format: Article
Language:English
Published: MDPI AG 2014-09-01
Series:Entropy
Subjects:
Online Access:http://www.mdpi.com/1099-4300/16/9/4855
_version_ 1811187056218472448
author Tarald O. Kvålseth
author_facet Tarald O. Kvålseth
author_sort Tarald O. Kvålseth
collection DOAJ
description Besides its importance in statistical physics and information theory, the Boltzmann-Shannon entropy S has become one of the most widely used and misused summary measures of various attributes (characteristics) in diverse fields of study. It has also been the subject of extensive and perhaps excessive generalizations. This paper introduces the concept and criteria for value validity as a means of determining if an entropy takes on values that reasonably reflect the attribute being measured and that permit different types of comparisons to be made for different probability distributions. While neither S nor its relative entropy equivalent S* meet the value-validity conditions, certain power functions of S and S* do to a considerable extent. No parametric generalization offers any advantage over S in this regard. A measure based on Euclidean distances between probability distributions is introduced as a potential entropy that does comply fully with the value-validity requirements and its statistical inference procedure is discussed.
first_indexed 2024-04-11T13:55:40Z
format Article
id doaj.art-1dde766676fa4496a007c3a08da78916
institution Directory Open Access Journal
issn 1099-4300
language English
last_indexed 2024-04-11T13:55:40Z
publishDate 2014-09-01
publisher MDPI AG
record_format Article
series Entropy
spelling doaj.art-1dde766676fa4496a007c3a08da789162022-12-22T04:20:20ZengMDPI AGEntropy1099-43002014-09-011694855487310.3390/e16094855e16094855Entropy Evaluation Based on Value ValidityTarald O. Kvålseth0Department of Mechanical Engineering, University of Minnesota, Minneapolis, MN 55455, USABesides its importance in statistical physics and information theory, the Boltzmann-Shannon entropy S has become one of the most widely used and misused summary measures of various attributes (characteristics) in diverse fields of study. It has also been the subject of extensive and perhaps excessive generalizations. This paper introduces the concept and criteria for value validity as a means of determining if an entropy takes on values that reasonably reflect the attribute being measured and that permit different types of comparisons to be made for different probability distributions. While neither S nor its relative entropy equivalent S* meet the value-validity conditions, certain power functions of S and S* do to a considerable extent. No parametric generalization offers any advantage over S in this regard. A measure based on Euclidean distances between probability distributions is introduced as a potential entropy that does comply fully with the value-validity requirements and its statistical inference procedure is discussed.http://www.mdpi.com/1099-4300/16/9/4855entropyrelative entropygeneralized entropiesentropy validityEuclidean entropy
spellingShingle Tarald O. Kvålseth
Entropy Evaluation Based on Value Validity
Entropy
entropy
relative entropy
generalized entropies
entropy validity
Euclidean entropy
title Entropy Evaluation Based on Value Validity
title_full Entropy Evaluation Based on Value Validity
title_fullStr Entropy Evaluation Based on Value Validity
title_full_unstemmed Entropy Evaluation Based on Value Validity
title_short Entropy Evaluation Based on Value Validity
title_sort entropy evaluation based on value validity
topic entropy
relative entropy
generalized entropies
entropy validity
Euclidean entropy
url http://www.mdpi.com/1099-4300/16/9/4855
work_keys_str_mv AT taraldokvalseth entropyevaluationbasedonvaluevalidity