The noncentral chi-squared distribution with noncentrality parameter is given by
|
(1)
| |||
|
(2)
| |||
|
(3)
|
where is a modified
Bessel function of the first kind and
is a regularized confluent hypergeometric limit
function. It is implemented in the Wolfram
Language as NoncentralChiSquareDistribution[r,
lambda].
For ,
its distribution function and survival function
can be written in terms of the generalized Marcum
Q-function as
|
(4)
| |||
|
(5)
|
(Peterson et al. 1995, pp. 672-681).
The mean, variance, skewness, and kurtosis excess are
|
(6)
| |||
|
(7)
| |||
|
(8)
| |||
|
(9)
|
The raw moments can be calculated analytically as
|
(10)
|
The first few are therefore
|
(11)
| |||
|
(12)
| |||
|
(13)
|
The first few central moments are
|
(14)
| |||
|
(15)
| |||
|
(16)
|