Home
Scholarly Works
Pitman closeness results for Type-I hybrid...
Journal article

Pitman closeness results for Type-I hybrid censored data from exponential distribution

Abstract

Recent work on Pitman closeness has compared estimators under Type-II censored samples from an exponential distribution. More recently, a comparison was done for Type-I censored data wherein two estimators based on two different censoring times were compared in their estimation of θ under Pitman closeness. Extending such a comparison, one may consider alternate comparisons under Type-I hybrid censoring. In this paper, we consider the comparison of three associated estimators based on three hybrid censored samples and carry out the analogous Pitman closeness comparisons. Formulas are derived for the suitable Pitman closeness probabilities and numerical results are tabulated for a variety of settings. While most of the tabulated probabilities agree with an intuition that the estimator based on larger termination time should be Pitman closer, exceptions are found.

Authors

Davies KF

Journal

Journal of Statistical Computation and Simulation, Vol. 91, No. 1, pp. 58–80

Publisher

Taylor & Francis

Publication Date

January 2, 2021

DOI

10.1080/00949655.2020.1806280

ISSN

0094-9655

Contact the Experts team