In statistics, Wilks' lambda distribution (named for Samuel S. Wilks), is a probability distribution used in multivariate hypothesis testing, especially with regard to the likelihood-ratio test and multivariate analysis of variance (MANOVA).
Wilks' lambda distribution is defined from two independent Wishart distributed variables as the ratio distribution of their determinants,[1]
given
A\simWp(\Sigma,m) B\simWp(\Sigma,n)
independent and with
m\gep
λ=
| \det(A) | |
| \det(A+B) |
=
| 1 | |
| \det(I+A-1B) |
\simΛ(p,m,n)
where p is the number of dimensions. In the context of likelihood-ratio tests m is typically the error degrees of freedom, and n is the hypothesis degrees of freedom, so that
n+m
There is a symmetry among the parameters of the Wilks distribution,[1]
Λ(p,m,n)\simΛ(n,m+n-p,p)
Computations or tables of the Wilks' distribution for higher dimensions are not readily available and one usually resorts to approximations.One approximation is attributed to M. S. Bartlett and works for large m[2] allows Wilks' lambda to be approximated with a chi-squared distribution
| \left( | p-n+1 |
| 2 |
-m\right)logΛ(p,m,n)\sim
| 2 | |
| \chi | |
| np |
.
Another approximation is attributed to C. R. Rao.[1] [3]
The distribution can be related to a product of independent beta-distributed random variables
ui\simB\left(
| m+i-p | , | |
| 2 |
| p | |
| 2 |
\right)
| n | |
| \prod | |
| i=1 |
ui\simΛ(p,m,n).
It follows directly that for a one-dimension problem, when the Wishart distributions are one-dimensional with
p=1
Λ(1,m,n)\simB\left(
| m | , | |
| 2 |
| n | |
| 2 |
\right).
From the relations between a beta and an F-distribution, Wilks' lambda can be related to the F-distribution when one of the parameters of the Wilks lambda distribution is either 1 or 2, e.g.,[1]
| 1-Λ(p,m,1) | |
| Λ(p,m,1) |
\sim
| p | |
| m-p+1 |
Fp,,
| 1-\sqrt{Λ(p,m,2) | |
\chi2