期刊名称：Sankhya. Series A, mathematical statistics and probability
印刷版ISSN：0976-836X
电子版ISSN：0976-8378
出版年度：2009
卷号：71
期号：01
页码：94--108
出版社：Indian Statistical Institute
摘要：This article deals with a reparametrized version of the inverse Gaussian dis-
tribution IG(; ). In this version we assume =
t and =
t2 for
; t > 0,
and note that t behaves like a convolution parameter. Next, we assume that
the conditional distribution of X given T = t is IG(
t;
t2) and examine
the bivariate distribution of (X; T) using various choices for the distribution
of T. In this context, with T random, we study Tweedie's original inverse
Gaussian independence results, and discuss extensions to the generalized in-
verse Gaussian case. Finally, we establish a characterization of the inverse
Gaussian distribution based on the bivariate distribution of (X; T).
