FrechetDistribution¶

represents a Frechet distribution.

$ wo 'FrechetDistribution[2, 3]'
FrechetDistribution[2, 3]

Mean and variance exist only for sufficiently large shape parameters (α > 1 and α > 2 respectively); outside those branches the result is Infinity:

$ wo 'Mean[FrechetDistribution[a, b]]'
Piecewise[{{b*Gamma[1 - a^(-1)], 1 < a}}, Infinity]

$ wo 'Mean[FrechetDistribution[2, 3]]'
3*Sqrt[Pi]

The median is β · Log[2]^(-1/α) for every parameter pair:

$ wo 'Median[FrechetDistribution[a, b]]'
b/Log[2]^a^(-1)