Bootstrapping Median Survival with Recurrent Event Data

In this paper, several resampling schemes to estimate the sampling distributions of median estimators of the inter-event time of a recurrent event are introduced and studied through simulations. Two types of recurrent event models are considered: first is a model where the inter-event times are independent and identically distributed, and second is a model where the inter-event times are associated, with the association arising from a gamma frailty model. The procedures studied are anchored on estimators proposed in Peña, Strawderman and Hollander (2001, Journal of the American Statistical Association 96, 1299-1315) and Wang and Chang (1999, Journal of the American Statistical Association 94, 146-153). The resampling procedures ∗email: [email protected] †email: [email protected]