The Analysis of Binary Longitudinal Data with Time-Dependent Covariates

• Activity Number: 175
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 10:30 AM to 12:20 PM
• Sponsor: ENAR
• Author(s): Matthew Guerra*+ and Justine Shults and Thomas R. Ten Have
• Companies: University of Pennsylvania and University of Pennsylvania and University of Pennsylvania School of Medicine
• Keywords: Alternating Logistic Regression ; Binary data ; Generalized Estimating Equations ; Longitudinal binary data ; Maximum Likelihood

Abstract

We consider longitudinal studies with binary outcomes that are measured repeatedly on subjects over time. Additional care must be taken to adjust for the association between the repeated measurements on each subject. We propose a new approach that extends the maximum likelihood method for time-independent covariates of Zeger, Liang, and Self (1985) to covariates that may be fixed or time-varying. We also implement and make comparisons with two other approaches: generalized estimating equations and alternating logistic regression. The proposed estimation procedure will yield consistent and asymptotically normal estimates of the regression and correlation parameters, if the correlation on consecutive measurements on a subject is constant. Simulations demonstrate that our approach can yield improved efficiency in estimation of the regression parameter.