Testing for the mean of the inverse gaussian distribution and adaptive estimation of parameters

Abstract

This paper provides the modified likelihood ratio criterion for testing whether the mean of the inverse Gaussian distribution can be set to unity giving rise to Standard form of the Wald distribution. Estimates of probability of correct selection has been obtained on the basis of a Monte Carlo study of 1,000 samples. Finally a set of adaptive estimators for the parameters are proposed and studied on the basis of data generated from the two distributions.

Keywords:

• Inverse Gaussian distribution
• Wald distribution
• Modified likelihood ratio
• Probability of correct selection
• Adaptive estimation