Event
Chien-Lin Su (Mark), PhD, Postdoctoral Fellow - McGill
Tuesday, February 13, 2018 15:30to16:30
Purvis Hall
Room 24, 1020 avenue des Pins Ouest, Montreal, QC, H3A 1A2, CA
The Accelerated Trend-Renewal Process for Recurrent Event Data.
Chien-Lin (Mark) Su is a postdoctoral fellowship at the department of Mathematics and Statistics, McGill University under the joint supervision of Professors Russell Steele and Johanna Nešlehová. He got his Ph.D. degree in statistics from Taiwan in 2015 and was awarded the STATLAB-CANSSI-CRM postdoc under the joint supervision of Professors Russell Steele (McGill University) and Lajmi Lakhal-Chaieb (Université Laval) from September 1st, 2016 to August 31, 2017. His research interests include multivariate survival analysis, copula model applications in biomedical research and causal inference.
Recurrent events data arises in many biomedical and longitudinal studies when failure events can occur repeatedly for each subject during the follow-up time. In this work, we are interested in the gap times between recurrent events. We propose a semiparametric accelerated transform gap time model based on trend-renewal process which contains a trend and a renewal component. We use the Buckley-James imputation approach to deal with censored transform gap times. The proposed estimators are shown to be consistent and asymptotically normal. Model diagnostic plots of residuals and a prediction method for predicting number of recurrent events given a specified covariate and follow-up time are also presented. Simulation studies are conducted to assess finite sample performances of the proposed method. The proposed technique is demonstrated through an application to two real data sets.
Recurrent events data arises in many biomedical and longitudinal studies when failure events can occur repeatedly for each subject during the follow-up time. In this work, we are interested in the gap times between recurrent events. We propose a semiparametric accelerated transform gap time model based on trend-renewal process which contains a trend and a renewal component. We use the Buckley-James imputation approach to deal with censored transform gap times. The proposed estimators are shown to be consistent and asymptotically normal. Model diagnostic plots of residuals and a prediction method for predicting number of recurrent events given a specified covariate and follow-up time are also presented. Simulation studies are conducted to assess finite sample performances of the proposed method. The proposed technique is demonstrated through an application to two real data sets.