Non-central Chi Square Distribution

The non-central chi square (χ2) distribution, also known as the generalized chi square distribution, is a generalization of the chi square distribution. It is used in the power analysis of statistical tests, including likelihood ratio tests.

Definition

The non-central chi square distribution with n degrees of freedom and non-centrality parameter λ is defined by its probability density function (PDF):

f(x;n,λ)=12e(x+λ)/2(xλ)(n2)/4I(n2)/2(λx)

where Iν(z) is the modified Bessel function of the first kind.

The parameters must satisfy n>0 and λ0.

Applications

  • Statistical power analysis uses it for chi-square and F-tests.

  • Signal processing employs it in radar detection theory.

  • Quality control uses it in process monitoring.

Properties

Statistical Properties
PropertyValue
Meann+λ
Variance2(2λ+n)
Skewness22(3λ+n)/(2λ+n)3/2
Excess Kurtosis12(4λ+n)/(2λ+n)2
Modeλ+n2 (for n2)

Notable properties include:

  • The characteristic function is ϕ(t)=(12it)n/2eλit/(12it).

  • The distribution is always right-skewed.

  • The moment generating function is M(t)=(12t)n/2eλt/(12t).

Relationships to Other Distributions

The NonCentralChiSquareDistribution class

The non-central chi square distribution is implemented by the NonCentralChiSquareDistribution class. It has one constructor which takes the degrees of freedom and the non-centrality parameter as arguments. The following constructs a non-central chi square distribution with 10 degrees of freedom and non-centrality parameter 15:

C#
var ncChiSquare = new NonCentralChiSquareDistribution(10, 15);

The NonCentralChiSquareDistribution class has two specific properties. DegreesOfFreedom returns the degrees of freedom of the distribution. NonCentralityParameter returns the non-centrality parameter.

References

  • Johnson, N. L., Kotz, S., & Balakrishnan, N. (1995). Continuous Univariate Distributions, Volume 1 (2nd ed.). Wiley.

  • Mood, A. M., Graybill, F. A., & Boes, D. C. (1974). Introduction to the Theory of Statistics (3rd ed.). McGraw-Hill.

See Also