The generalized inverse Gaussian distribution is the continuous
distribution on
with probability
density function
where ,
is real, and
is the modified
Bessel function of the second kind (Jørgensen 1982). The case
is an inverse
Gaussian distribution after parameter conversion.
Mao and Tian (2026) proved that, if and
,
then a
random variable
has the distribution of
,
where
has distribution
and
is independent. Their analytic criterion is that
is a completely monotonic function
on
in the stated parameter range. The lower endpoint
is not included in this statement.