|
Abstract:
|
In survival analysis it often happens that some subjects under study do not experience the event of interest; they are considered to be cured. The population is thus a mixture of two subpopulations: the one of cured subjects, and the one of susceptible subjects. Based on inverse probability censoring weighting, we propose new regression approaches for the cure rate, that is for the conditional probability of being cured, when the data are subject to random right censoring. Both continuous and discrete covariates are allowed. The parameters of interest could be estimated by least squares or maximum likelihood, without any reference to the law of the susceptible subjects. Confidence regions could be easily built by multiplier bootstrap. Penalized versions of the estimation criteria allow for variable selection. A new model check procedure that avoids the curse of dimensionality is also proposed. The asymptotic results and the inference are based on the iid representation of the conditional survival function estimator of the censoring variable. Such representations could be derived under mild conditions. Simulation experiments illustrate the effectiveness of the new approaches.
|
