Polar transformation based Student T random number generator.
See "Polar Generation of Random Variates With the t-Distribution",
Ralph W. Bailey, April 1994, in Mathematics of Computation, Vol 62-206
page 779.
The one implemented here is a variant of this algorithm from "Random
Number Generation and Monte Carlo Methods", Springer, 2003, page 185.
Using a uniform RNG on a [-1,1] support, the extra call to the uniform
generator (used for the sign of the sample) is avoided.
Class RNG must implement the following interface:
URNG::sample_type RNG::next() const;
warning
do not use with a low-discrepancy sequence generator.
Student t random number generator
Polar transformation based Student T random number generator. See "Polar Generation of Random Variates With the t-Distribution", Ralph W. Bailey, April 1994, in Mathematics of Computation, Vol 62-206 page 779. The one implemented here is a variant of this algorithm from "Random Number Generation and Monte Carlo Methods", Springer, 2003, page 185. Using a uniform RNG on a [-1,1] support, the extra call to the uniform generator (used for the sign of the sample) is avoided.
Class RNG must implement the following interface:
URNG::sample_type RNG::next() const;do not use with a low-discrepancy sequence generator.