Stochastic monotonicity of the MLE of exponential mean

under different censoring schemes

N. Balakrishnan and G. Iliopoulos

In this paper, we present a general method which can be used in order to show that the maximum likelihood estimator (MLE) of an exponential mean $\theta$ is stochastically increasing with respect to $\theta$ under different censored sampling schemes. This propery is essential for the construction of exact confidence intervals for $\theta$ via ``pivoting the cdf'' as well as for the tests of hypotheses about $\theta$. The method is shown for Type-I censoring, hybrid censoring and generalized hybrid censoring schemes.  We also establish the result for the exponential competing risks model with censoring.

Key words and phrases: Exponential distribution, maximum likelihood estimation, Type-I censoring, Type-I and Type-II hybrid censoring, Type-I and Type-II generalized hybrid censoring, exact confidence intervals, stochastic ordering, competing risks model.

Download (184Kb) (The original publication is available at www.springerlink.com)

Back to Home Page