A Nonparametric Quantile Estimator: Computation.
SOUTH CAROLINA UNIV COLUMBIA DEPT OF STATISTICS
Pagination or Media Count:
Right-censored data arise very naturally in life testing and reliability studies. For such data, it is important to be able to obtain good nonparametric estimates of various characteristics of the unknown lifetime distribution. This report concerns the computational procedure for a kernel-type nonparametric estimator of the quantile function of the lifetimne distribution from right-censored data. This estimator was suggested by Padgett 1986, extending the complete sample results of Yang 1985. The large sample properties of the estimator, such as asymptotic normality and mean square convergence, were studied by Lio, Padgett and Yu 1986 and by Lio and Padgett 1985. In this report, a procedure for calculation of the kernel-type quantile estimate from right-censored data is described, and a listing of a computer program in FORTRAN code is provided.
- Statistics and Probability