Proceedings of the Second Seattle Symposium in Biostatistics: Analysis of Correlated Data (Lecture Notes in Statistics, 179)

This volume contains a selection of papers presented at the Second Seattle Symposium in Biostatistics: Analysis of Correlated Data. The symposium was held in 2000 to celebrate the 30 th anniversary of the University of Washington School of Public Health and Community Medicine. It featured keynote lectures by Norman Breslow, David Cox and Ross Prentice and 16 invited presentations by other prominent researchers. The papers contained in this volume encompass recent methodological advances in several important areas, such as longitudinal data, multivariate failure time data and genetic data, as well as innovative applications of the existing theory and methods. This volume is a valuable reference for researchers and practitioners in the field of correlated data analysis. Front Matter....Pages i-vii Whither PQL?....Pages 1-22 Correlation and Marginal Longitudinal Kernel Nonparametric Regression....Pages 23-33 Analysis of Multivariate Monotone Missing Data by A Pseudolikelihood Method....Pages 35-50 Quantile Regression for Correlated Observations....Pages 51-69 Small Sample Inference for Clustered Data....Pages 71-87 Some Applications of Indirect Inference to Longitudinal and Repeated Events Data....Pages 89-105 On Characterizing Joint Survivor Functions by Minima....Pages 107-111 Nonparametric Estimation of the Bivariate Survivor Function....Pages 113-142 A Semiparametric Regression Model for Panel Count Data: When Do Pseudo-likelihood Estimators Become Badly Inefficient?....Pages 143-174 Some Biases That May Affect Kin-Cohort Studies for Estimating the Risks from Identified Disease Genes....Pages 175-187 Optimal Structural Nested Models for Optimal Sequential Decisions....Pages 189-326 Back Matter....Pages 327-331 "The papers contained in this volume encompass recent methodological advances in several important areas, such as longitudinal data, multivariate failure time data and genetic data, as well as innovative applications of the existing theory and methods. This volume is a valuable reference for researchers and practitioners in the field of correlated data analysis."--Jacket Penalized Quasi-Likelihood is a technique for approximate inference in GLMMs and is not a rigorous statistical method in its own right.