Estimating powers of the scale parameters under order restriction for two shifted exponential populations with a common location.

This article investigates the problem of estimating the powers of scale parameters under order restriction of two shifted exponential populations when the location parameters are assumed to be unknown and equal. Several classical estimators have been proposed, such as the maximum likelihood estimators, plug-in type restricted maximum likelihood estimators, and the uniform minimum variance unbiased estimators. Sufficient conditions for constructing improved estimators under the scale and affine group of transformations have been derived. Consequently, several improved estimators for the powers of the scale parameters under order restriction have been proposed. Furthermore, using the quadratic loss function, a simulation study has been carried out to compare all the proposed estimators in terms of risk values, and recommendations are made there. [ABSTRACT FROM AUTHOR]