摘要：In this paper, we propose a exible cure rate survival model by assumingthat the number of competing causes of the event of interest followsthe negative binomial distribution and the time to event follows a generalizedgamma distribution. We de ne the negative binomial-generalized gammadistribution, which can be used to model survival data. The new model includesas special cases some of the well-known cure rate models discussed inthe literature. We consider a frequentist analysis and nonparametric bootstrapfor parameter estimation of a negative binomial-generalized gammaregression model with cure rate. Then, we derive the appropriate matricesfor assessing local inuence on the parameter estimates under di erentperturbation schemes and present some ways to perform global inuenceanalysis. Finally, we analyze a real data set from the medical area.
