Gets the probability for the ratio of two variances from the selected Covariance object being different from a hypothesized ratio.
The test statistic
|f = s12 / s22 / ratio|
is distributed as Fisher's F distribution with ndf1 = N-1 and ndf2 = N-1 degrees of freedom for the numerator and denominator terms, respectively.
The returned probability p will be the two-sided probability
|p = 2 * fisherQ (f, ndf1, ndf2)|
If s22 > s12 we use 1/f to determine the probability.
© djmw, April 7, 2004