Analysis For Multiple Type Recurrent Events Data

In many application research region such as biostatistics, clinical medicine and industry re-liability, individuals may repeat experience of a certain event or failure, which was known asrecurrent events. Recurrent events data refers to the observation of individuals, which contain therecurrent event time of interest. When focus on single type event, single type recurrent event dataare obtained. If we are interested in several types of recurrent events, and those result may occurseveral times within observation time, multi-type recurrent events are obtained.For multi-type recurrent events, the effort of covariate that effect recurrent events are dis-cussed in detail. The thesis make a briefly introduction about survival function, proportional haz-ard model and recurrent events statistical model. Our attention is dedicate to additive-acceleratedregression means model based on multi-type recurrent events, methods for inferences on thismodel are proposed. Under modern empirical theory, the estimator asymptotic properties areobtained. For additive-accelerated hazard model based on multi-type recurrent events gap time,methods for inferences on the model are presented. Regularity conditions is given to discuss theestimator consistency and asymptotic properties.