An information theoretical algorithm for analyzing supersaturated designs for a binary response

Balakrishnan, N; Koukouvinos, C; Parpoula, C

dc.contributor.author	Balakrishnan, N	en
dc.contributor.author	Koukouvinos, C	en
dc.contributor.author	Parpoula, C	en
dc.date.accessioned	2014-03-01T11:46:27Z
dc.date.available	2014-03-01T11:46:27Z
dc.date.issued	2011	en
dc.identifier.issn	00261335	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/37925
dc.subject	Entropy	en
dc.subject	Error rates	en
dc.subject	Factor screening	en
dc.subject	Generalized linear models	en
dc.subject	Information gain	en
dc.subject	ROC	en
dc.subject	Symmetrical uncertainty	en
dc.title	An information theoretical algorithm for analyzing supersaturated designs for a binary response	en
heal.type	other	en
heal.identifier.primary	10.1007/s00184-011-0373-5	en
heal.identifier.secondary	http://dx.doi.org/10.1007/s00184-011-0373-5	en
heal.publicationDate	2011	en
heal.abstract	A supersaturated design is a factorial design in which the number of effects to be estimated is greater than the number of runs. It is used in many experiments, for screening purpose, i.e., for studying a large number of factors and identifying the active ones. In this paper, we propose a method for screening out the important factors from a large set of potentially active variables through the symmetrical uncertainty measure combined with the information gain measure. We develop an information theoretical analysis method by using Shannon and some other entropy measures such as Rényi entropy, Havrda-Charvát entropy, and Tsallis entropy, on data and assuming generalized linear models for a Bernoulli response. This method is quite advantageous as it enables us to use supersaturated designs for analyzing data on generalized linear models. Empirical study demonstrates that this method performs well giving low Type I and Type II error rates for any entropy measure we use. Moreover, the proposed method is more efficient when compared to the existing ROC methodology of identifying the significant factors for a dichotomous response in terms of error rates. © 2011 Springer-Verlag.	en
heal.journalName	Metrika	en
dc.identifier.doi	10.1007/s00184-011-0373-5	en
dc.identifier.spage	1	en
dc.identifier.epage	18	en