SMOOTH NONPARAMETRIC ESTIMATION OF MEAN RESIDUAL LIFE

  • Na, Myoung-Hwan (Department of Statistics, Seoul National University) ;
  • Park, Sung-Hyun (Department of Statistics, Seoul National University) ;
  • Kim, Jae-Joo (Department of Statistics, Seoul National University)
  • Published : 1998.11.01

Abstract

In this paper we propose smooth nonparametric estimator of Mean Residual Life(MRL) based on a complete sample. This estimator is constructed using estimator of cumulative failure rate which is derived as the maximum likelihood estimate of cumulative failure rate in the class of distributions which have piecewise linear failure rate functions between each pair of observations. We derive the asymptotic properties of the our estimator. The proposed estimator is compared with previously known estimator by Monte Carlo study.

Keywords