國立臺灣大學數學系CHIN-TSANG CHIANGWang, Mei-ChengMei-ChengWangChiang, Chin-TsangChin-TsangChiang2006-09-272018-06-282006-09-272018-06-282002http://ntur.lib.ntu.edu.tw//handle/246246/20060927121125210253Recurrent event data are commonly encountered in health-related longitudinal studies. In this paper timeto- events models for recurrent event data are studied with non-informative and informative censorings. In statistical literature, the risk set methods have been con5rmed to serve as an appropriate and e6cient approach for analysing recurrent event data when censoring is non-informative. This approach produces biased results, however, when censoring is informative for the time-to-events outcome data. We compare the risk set methods with alternative non-parametric approaches which are robust subject to informative censoring. In particular, non-parametric procedures for the estimation of the cumulative occurrence rate function (CORF) and the occurrence rate function (ORF) are discussed in detail. Simulation and an analysis of data from the AIDS Link to Intravenous Experiences Cohort Study is presented.application/pdf337442 bytesapplication/pdfzh-TWcumulative rate functioninformative censoringintensity functionkernal estimationrate functionrecurrent events[SDGs]SDG3[SDGs]SDG16acquired immune deficiency syndrome; article; cohort analysis; confidentiality; controlled study; data analysis; human; intravenous drug abuse; longitudinal study; major clinical study; outcomes research; patient information; recurrence risk; risk assessment; simulation; statistical analysis; time; Acquired Immunodeficiency Syndrome; Computer Simulation; Hospitalization; Humans; Longitudinal Studies; Models, Statistical; Monte Carlo Method; Statistics, Nonparametric; Substance Abuse, Intravenousjournal article10.1002/sim.10292-s2.0-0037083157WOS:000173737300011http://ntur.lib.ntu.edu.tw/bitstream/246246/20060927121125210253/1/fulltext.pdf