Abstract

This article is concerned with the loss of information that arises due to Type I censoring while discriminating between two Pareto populations. The discrepancy criterion used is due to Kullback and Leibler. It is shown that the loss of information depends on the truncation time as specified by the censoring procedure. An expression for the residual duration of the experiment is also obtained. Finally, a numerical illustration with simulated data is given.

Author: A. B. Aich

Received on: March, 2005

Accepted on: January, 2006