| Title: | SECOND ORDER ESTIMATING EQUATIONS FOR CLUSTERED LONGITUDINAL BINARY DATA WITH MISSING OBSERVATIONS |
| DOI No: | 10.1142/9781860949531_0031 |
| Source: | RECENT ADVANCES IN STATISTICAL METHODS (pp 352-366)
|
| Author(s): | GRACE Y. YI
Department of Statistics and Actuarial Science, University of Waterloo, 200 University Avenue West, Waterloo, Ontario, Canada, N2L 3G1, Canada
RICHARD J. COOK
Department of Statistics and Actuarial Science, University of Waterloo, 200 University Avenue West, Waterloo, Ontario, Canada, N2L 3G1, Canada
|
| Abstract: | For incomplete longitudinal data Robins et al. l5 developed inverse probability weighted generalized estimating equations for the marginal mean parameters. In many cases, however, the repeated measurements themselves may arise in clusters, which leads to both a cross-sectional and a longitudinal correlation structure. In some applications the correlation structure may become of interest itself. In this paper we describe second order inverse probability weighted generalized estimating equations for association parameters characterizing the dependence among observations within clusters. The inverse probability weights are estimated from conditional logistic models for the missing data process. The methods are applied to data from the Waterloo Smoking Prevention Project for illustrative purposes. |
| Full Text: | View full text in PDF format (715KB) |
| TOC: | Back to Table of Contents |
|
|
