Please use this identifier to cite or link to this item:
DC FieldValue
dc.titleInfeasibility and efficiency of working correlation matrix in generalized estimating equations
dc.contributor.authorNG WEE TECK
dc.identifier.citationNG WEE TECK (2006-02-06). Infeasibility and efficiency of working correlation matrix in generalized estimating equations. ScholarBank@NUS Repository.
dc.description.abstractGeneralized estimating equations (GEE) have been used extensively in the analysis of clustered and longitudinal data since the seminal paper by Liang & Zeger (1986). A key attraction of using GEE is that regression parameters remain consistent even when the 'working' correlation is misspecified. However, Crowder (1995) pointed out that there are problems with the estimation of the correlation parameters when it is misspecified and this affects the consistency of the regression parameters. This issue has been addressed by Chaganty & Shults (1999), however the estimators are asymptotically biased. The feasibility of the estimators for the correlation parameters under misspecification efficiency of the various methods of estimating the correlation parameters under misspecification are investigated in this thesis. Analytic expressions for the estimating functions using the decoupled Gaussian and Cholesky decomposition proposed by Wang & Carey (2004) are for common correlation structures such as exchangeable, AR(1) and MA(1) are also provided.
dc.subjectEfficiency; Estimating function; Longitudinal data; CORRELATION; Repeated measures; Pseudolikelihood
dc.contributor.departmentSTATISTICS & APPLIED PROBABILITY
dc.contributor.supervisorWANG YOUGAN
dc.description.degreeconferredMASTER OF SCIENCE
Appears in Collections:Master's Theses (Open)

Show simple item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
NgWT.pdf326.05 kBAdobe PDF



Page view(s)

checked on May 23, 2019


checked on May 23, 2019

Google ScholarTM


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