Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/13704
Title: A solution to separation for clustered binary data
Authors: Abrahantes, Jose Cortinas
AERTS, Marc 
Issue Date: 2012
Publisher: SAGE PUBLICATIONS LTD
Source: STATISTICAL MODELLING, 12 (1), p. 3-27
Abstract: The presence of one or more covariates that perfectly or almost perfectly predict the outcome of interest (which is referred to as complete or quasi-complete separation, the latter denoting the case when such perfect prediction occurs only for a subset of observations in the data) has been extensively studied in the last four decades. Since 1984, when Albert and Anderson (1984) differentiated between complete and quasi-complete separation, several authors have studied this phenomenon and tried to provide answers or ways of identifying the problem (Lesaffre and Albert, 1989; Firth, 1993; Christmann and Rousseeuw, 2001; Rousseeuw and Christmann, 2003; Allison, 2004; Zorn, 2005; Heinze, 2006). From an estimation perspective, separation leads to infinite coefficients and standard errors, which makes the algorithm collapse or give inappropriate results. As a practical matter, separation forces the analyst to choose from a number of problematic alternatives for dealing with the problem, and in the past the elimination of such problematic variables were common practice to deal with such situations. In the last decade, solutions using penalized likelihood have been proposed, but always dealing with independent binary data. Here we will propose a Bayesian solution to the problem when we deal with clustered binary data using informative priors that are supported by the data and compare it with an alternative procedure proposed by Gelman et al. (2008).
Notes: [Abrahantes, Jose Cortinas] European Food Safety Author EFSA, I-43126 Parma, Italy. [Abrahantes, Jose Cortinas; Aerts, Marc] Hasselt Univ Belgium, Ctr Stat, Interuniv Inst Biostat & Stat Bioinformat I BioSt, Hasselt, Belgium. jose.cortinasabrahantes@efsa.europa.eu
Keywords: Statistics & Probability; Separation issues; clustered binary data; logistic model; Bayesian analysis; conditional models; penalized likelihood approach;Separation issues; clustered binary data; logistic model; Bayesian analysis; conditional models; penalized likelihood approach
Document URI: http://hdl.handle.net/1942/13704
ISSN: 1471-082X
e-ISSN: 1477-0342
DOI: 10.1177/1471082X1001200102
ISI #: 000302437700002
Category: A1
Type: Journal Contribution
Validations: ecoom 2013
Appears in Collections:Research publications

Files in This Item:
File Description SizeFormat 
Cortinas and Aerts 2012.pdfarticle135.62 kBAdobe PDFView/Open
Show full item record

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.