Semiparametric Estimation with Correlated Recurrent Event Data under Informative Monitoring

Author

Akim Adekpedjou, Missouri University of Science and TechnologyFollow
Jonathan Quiton

Abstract

Consider a recurrent event data where frailty models are used to account for correlations among the inter-event times within each unit under study. In this talk we consider the problem of semiparametric estimation of the inter-event time distribution under informative monitoring and the Gamma frailty model. We propose a semiparametric estimator of the baseline survivor function. We show that the estimator under the i.i.d. setting is inconsistent in the presence of frailty. We present results of simulaiton study where we discuss the performence of our proposed estimator to that derived under the i.i.d. setting. Finally, these estimators will be demonstrated by applying to biomedical and engineering data sets.

Recommended Citation

A. Adekpedjou and J. Quiton, "Semiparametric Estimation with Correlated Recurrent Event Data under Informative Monitoring," Proceedings of the 2008 Joint Statistical Meetings (2008, Denver, CO), American Statistical Association, Aug 2008.

Meeting Name

2008 Joint Statistical Meetings (2008: Aug. 3-7, Denver, CO)

Department(s)

Mathematics and Statistics

Keywords and Phrases

Correlated Recurrent times; Frailty model; Informative Censoring; Koziol-Green Model

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2008 American Statistical Association, All rights reserved.

Publication Date

07 Aug 2008