Improvement of Kernel Estimators of the Failure Rate Function Using the Generalized Jacknife.
Abstract
In this paper we explore methods by which the rate of convergence of the bias and the mean square error of kernel estimators of the failure rate function can be improved. We show that if the kernel is not restricted to be nonnegative, and is suitably chosen, then the bias contribution to the asymptotic mean square error can be eliminated to any required order, and the rate of convergence of the asymptotic mean square error can be brought as close to 1/n as is desired. The generalized jackknife method of combining estimators is shown to be an adequate procedure which leads us to this goal. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 14, 1980
- Accession Number
- ADA092928
Entities
People
- Man-yuen Wong
- Nozer Singpurwalla
Organizations
- George Washington University