Fisher distribution with the number of degrees of freedom d1 and d2 is described by probability density function:
Where B - a beta function which looks as follows:
If two independent random values are described with χ2 distribution Y1 ~ χ2(d1) and Y2 ~ χ2(d2), the random value
has Fisher distribution with the degrees of freedom (d 1, d 2). As χ 2 (d i ) distribution is a special case of gamma distribution, it matches Г (d i/2, 2). The following steps are executed:
1. Using gamma distribution, generate two independent pseudo-random values Y1 ~ Г(d1/2, 2) and Y2 ~ Г(d2/2, 2).
2. Values of target sample elements are calculated:
See also: