Inference on overlap in two inverse Gaussian populations
by
Yogendra P. Chaubey
Concordia University
Coauthors: Debaraj Sen (Concodia University) and Satya N. Mishra (University of South Alabama)

The inverse Gaussian distribution is often suited for modelling positive and/or positively skewed data [see Chhikara and Folks (1989)] and presents an interesting alternative to the Gaussian model in such cases. We note here that overlap coefficients and their variants are widely studied in the literature for Gaussian populations [see Mulekar and Mishra (1994, 2000) and references there in for further details]. This paper addresses point and interval estimation of commonly used measures of overlap when the populations are described by inverse class of distributions.It is found that for one of the measures, namely the Matusita’s measure, the inference procedures developed in Mulekar and Mishra (1994, 2000) carry over while for other measures, they do not.

Date received: August 16, 2004