Adaptive Goodness-of-Fit Tests, Recurrent Event Models, and Models with Alternative Time-Scales

University Of South Carolina At Columbia, Columbia SC

Investigators

Abstract

Abstract DMS 0102870 Edsel Pena This project will address statistical research problems in hazard-based models in the face of incomplete data and optimal adaptive tests for these models, by investigating conditions for their existence, construction and application. Additionally, stochastic problems in recurrent events applications will be considered. A new and general class of recurrent events analyses that will encompass the special cases now in the literature is proposed. This class features models that simultaneously incorporate interventions as they occur and account for the effects of concomitant variables. The major contributions of this project are expected to enhance the theoretical understanding of inferences in a variety of hazard-based models. Hazard-based models have wide application. They are used to understand reliability of many operations, including industrial, biomedical and economic. The procedures to be developed under this grant are primarily for the purpose of understanding the fundamental aspects of these experiences and experiments. These models may model rare events such as flooding and industrial accidents, but also be appropriate for modeling hospital admissions and stock market cycles. Thus, these statistical methods very useful for further research. The Principal Investigator will enable many disadvantaged students from a large number of disciplines to participate in his education/research efforts.