Estimation of the Parameters of the Uniform Distribution Over [kθ, (k+ 1)θ]

From the available literature on estimation of the parameters of the uniform distribution over [ kθ, kθ + θ], we find the necessity to construct improved estimators of the parameter θ when k is known. In this paper a new estimator of θ is proposed when k is known and its performance is compared with best linear unbiased estimator of θ based on two extreme observations. A MS Subject Classification: Primary: 62H12; Secondary: 62G30, 62B05.

A Class of Tests for the Dispersive-Equivalence of two Probability Distributions

Based on independent random samples from two absolutely continuous distributions, asymptotically distribution-free and consistent tests are proposed for testing that two distributions are identical except for an unknown location parameter against the alternative that one is dispersive ordered with respect to the other. These tests are based on weighted combinations of ratios of spacings. AMS 1980 Subject Classification: Primary 62Gl0, secondary 62G30.