Resumen de A fexible marginal rate model for recurrent events with a zero-recurrence proportion

Ivo Sousa-Ferreira, Cristina Rocha, Ana Maria Abreu

• A new marginal rate model for gap times between recurrent events is proposed, which is derived from a non-homogeneous Poisson process (NHPP).

  Since the distribution of the gap times often requires exible shapes of the rate function, the approach taken here is to model the baseline log-cumulative rate function as a restricted cubic spline function of log time. Moreover, a proportion of subjects that will never experience any recurrence is incorporated. The proposed model allows covariates in both the latency and incidence components. An application to a real data set is also provided for illustrative purposes.