Nonparametric hazard estimation with time-varying discrete covariates
含时变离散协变量的非参数风险函数估计
The traditional one-sample nonparametric estimator of the cumulative hazard function is extended to allow for time-varying discrete covariates. A proof for the consistency and asymptotic normality of the estimator, as well as a derivation of the asymptotic standard error, is provided. Numerical results show that the estimator performs reasonably well.