Nonparametric Regression for Event Times in Multistate Models with Clustered Current Status Data with Informative Cluster Size

• Activity Number: 603
• Type: Contributed
• Date/Time: Wednesday, August 7, 2013 : 2:00 PM to 3:50 PM
• Sponsor: ENAR
• Author(s): Ling Lan*+ and Dipankar Bandyopadhyay and Somnath Datta
• Companies: Georgia Regents University and University of Minnesota and University of Louisville
• Keywords: Multi-state model ; informative cluster size ; current status data ; nonparametric ; multivariate survival analysis ; dental data

Abstract

We consider data sets containing current status information on individual units each undergoing a multistate system. The state processes have a clustering structure such that the size of each cluster is random and may be in correlation with common characteristics of the multistate models in the cluster. We propose nonparametric estimators for the state occupation probabilities at a given time conditional on a continuous and a discrete covariate. Weighted monotonic regression and smoothing are used to uniquely define the state occupation probability regression estimators. A detailed simulation study evaluated the global performance of the proposed nonparametric estimator. An illustrative application to a dental disease data is also presented.