Analyzing supersaturated designs with entropic measures

C Koukouvinos; E. Massou; Kalliopi Mylona; Christina Parpoula

doi:10.1016/j.jspi.2010.10.001

Analyzing supersaturated designs with entropic measures

C Koukouvinos, E. Massou, Kalliopi Mylona, Christina Parpoula

National Technical University of Athens

Research output: Contribution to journal › Article › peer-review

4 Citations (Scopus)

Abstract

A supersaturated design is a design for which there are fewer runs than effects to be estimated. In this paper, we propose a method for screening out the important factors from a large set of potentially active variables, based on an information theoretical approach. Three entropy measures: Re´nyi entropy, Tsallis entropy and Havrda–Charva´t entropy, have been associated with the measure of information gain, in order to identify the significant factors using data and assuming generalized linear models. The investigation of the proposed method performance and the comparison of each entropic measure application have been accomplished through simulation experiments. A noteworthy advantage of this paper is the use of generalized linear models for analyzing data from supersaturated designs, a fact that, to the best of our knowledge, has not yet been studied.

Original language	English
Pages (from-to)	1307–1312
Journal	JOURNAL OF STATISTICAL PLANNING AND INFERENCE
Volume	141
Issue number	3
Early online date	17 Oct 2010
DOIs	https://doi.org/10.1016/j.jspi.2010.10.001
Publication status	E-pub ahead of print - 17 Oct 2010

Access to Document

10.1016/j.jspi.2010.10.001

Cite this

@article{256d2e6b94a841bfb9bc532a9de61321,

title = "Analyzing supersaturated designs with entropic measures",

abstract = "A supersaturated design is a design for which there are fewer runs than effects to be estimated. In this paper, we propose a method for screening out the important factors from a large set of potentially active variables, based on an information theoretical approach. Three entropy measures: Re´nyi entropy, Tsallis entropy and Havrda–Charva´t entropy, have been associated with the measure of information gain, in order to identify the significant factors using data and assuming generalized linear models. The investigation of the proposed method performance and the comparison of each entropic measure application have been accomplished through simulation experiments. A noteworthy advantage of this paper is the use of generalized linear models for analyzing data from supersaturated designs, a fact that, to the best of our knowledge, has not yet been studied.",

author = "C Koukouvinos and E. Massou and Kalliopi Mylona and Christina Parpoula",

year = "2010",

month = oct,

day = "17",

doi = "10.1016/j.jspi.2010.10.001",

language = "English",

volume = "141",

pages = "1307–1312",

journal = "JOURNAL OF STATISTICAL PLANNING AND INFERENCE",

issn = "0378-3758",

publisher = "Elsevier",

number = "3",

}

TY - JOUR

T1 - Analyzing supersaturated designs with entropic measures

AU - Koukouvinos, C

AU - Massou, E.

AU - Mylona, Kalliopi

AU - Parpoula, Christina

PY - 2010/10/17

Y1 - 2010/10/17

N2 - A supersaturated design is a design for which there are fewer runs than effects to be estimated. In this paper, we propose a method for screening out the important factors from a large set of potentially active variables, based on an information theoretical approach. Three entropy measures: Re´nyi entropy, Tsallis entropy and Havrda–Charva´t entropy, have been associated with the measure of information gain, in order to identify the significant factors using data and assuming generalized linear models. The investigation of the proposed method performance and the comparison of each entropic measure application have been accomplished through simulation experiments. A noteworthy advantage of this paper is the use of generalized linear models for analyzing data from supersaturated designs, a fact that, to the best of our knowledge, has not yet been studied.

AB - A supersaturated design is a design for which there are fewer runs than effects to be estimated. In this paper, we propose a method for screening out the important factors from a large set of potentially active variables, based on an information theoretical approach. Three entropy measures: Re´nyi entropy, Tsallis entropy and Havrda–Charva´t entropy, have been associated with the measure of information gain, in order to identify the significant factors using data and assuming generalized linear models. The investigation of the proposed method performance and the comparison of each entropic measure application have been accomplished through simulation experiments. A noteworthy advantage of this paper is the use of generalized linear models for analyzing data from supersaturated designs, a fact that, to the best of our knowledge, has not yet been studied.

U2 - 10.1016/j.jspi.2010.10.001

DO - 10.1016/j.jspi.2010.10.001

M3 - Article

SN - 0378-3758

VL - 141

SP - 1307

EP - 1312

JO - JOURNAL OF STATISTICAL PLANNING AND INFERENCE

JF - JOURNAL OF STATISTICAL PLANNING AND INFERENCE

IS - 3

ER -

Analyzing supersaturated designs with entropic measures

Abstract

Access to Document

Fingerprint

Cite this