|
Abstract:
|
Multi-type recurrent events are commonly observed in commercial truck transportation studies, since the drivers may encounter different types of safety critical events and take different lengths of on-duty breaks in a driving shift. Bayesian non-homogeneous Poisson process models are a flexible approach to jointly model the intensity functions of the multi-type recurrent events. For comparing these models, the deviance information criterion and the logarithm of the pseudo-marginal likelihood are studied, and Monte Carlo methods are developed for computing these model assessment measures. We also propose a set of new concordance indices (C-indices) to evaluate various discrimination abilities of a Bayesian multi-type recurrent event model. Specifically, the within-event C-index, the between-event C-index and the overall C-index quantify adequacies of a given model in fitting the recurrent event data for each type, between two types, and for all of the types together, respectively. Moreover, an in-depth analysis of a real data set from the commercial truck driver naturalistic driving study is carried out to demonstrate the usefulness and applicability of the proposed methodology.
|
