Survival-rate regression using kernel conditional Kaplan-Meier estimators

We consider a regression model in which it is assumed that the conditional survival distribution of the response given the covariate, after transformation using a link function, satisfies a linear regression model. By proper choice of the link function the logistic and Cox models can be obtained. The response is allowed to be subject to right censoring. We consider an estimation procedure for the regression parameters and establish the asymptotic normality of the estimator when the covariate is one-dimensional. The finite sample performance of the proposed estimator is studied through simulations.