An EM algorithm for the destructive COM-Poisson regression cure rate model (Q682051)

This paper deals with right censored survival data when the initial number of competing causes undergoes destruction after an initial treatment. Assuming a Conway-Maxwell Poisson distribution for the initial risk factors function and a Weibull distribution to the time-to-event, maximum likelihood estimation of the survival model parameters by means of an EM algorithm is proposed. The estimation of parameters of an explanatory model for the cure rates is also considered. A Monte Carlo simulation study is carried out to evaluate the performance of the proposed EM algorithm. An application of the estimation method to a real data set is also presented.