non-central chi-squared random variable
Let be IID random variables, each with the standard normal distribution
. Then, for any , the random variable defined by
is called a non-central chi-squared random variable.Its distribution depends only on the number of degrees of freedom and non-centrality parameter . This is denoted by and has moment generating function
(1) |
which is defined for all with real part less than .More generally, for any , not necessarily integers, a random variable has the non-central chi-squared distribution, , if its moment generating function is given by (1).
A non-central chi-squared random variable for any can be constructed as follows. Let be a (central) chi-squared variable with degree , be standard normals, and have the distribution. If these are all independent then
has the distribution. Correspondingly, the probability density function for is
(2) |
where and is the probability density of the distribution.Alternatively, this can be expressed as
where is a modified Bessel function of the first kind,
Remarks
- 1.
has mean and variance
.
- 2.
. The (central) chi-squared random variable is a special case of the non-central chi-squared random variable, when the non-centrality parameter .
- 3.
(The reproductive property of chi-squared distributions). If are non-central chi-squared random variables such that each , then their total is also a non-central chi-squared random variable with distribution .
- 4.
If then the distribution is restricted to the domain with probability density function (2). On the other hand, if , then there is also an atom at ,
- 5.
If is a multivariate normally distributed -dimensional random vector with distribution where is the mean vector and is the covariance matrix
. Suppose that is singular
, with = rank of . Then is a non-central chi-squared random variable, where is a generalized inverse of . Its distribution has degrees of freedom with non-centrality parameter .