Efficiency of schemes of sampling from the inverse gaussian-weibull mixture model

Abstract

The objective of this paper is to study the efficiency of sampling schemes suggested by Hosmer(1973), termed models Ml and M2, relative to the regular random sampling, termed model MO, when samples are drawn from a population having the Inverse Gaussian-Weibull (IG-W) mixture distribution.

It has been shown that whether the efficiency is based on relative variances of the maximum likelihood estimates (ML,E's) of the components of the vector of parameters or on the generalized variances of the MLE's of that vector, Hosmer's models Ml or M2 perform better than model MO.

Keywords:

• inverse gaussian and Weibull distnibutions
• mixture
• Hosmer's schemes of sampling
• information matrix
• asynptatic variance-cavariance matrix
• generalized variance
• efficiency