On Mon, Feb 1, 2010 at 5:59 PM, Steven Watanabe
AMDG
Marco Guazzone wrote:
But for what regards Gamma distribution I would to point out that actually the code used to generate random Gamma variates cannot be used to generate variates for two-parameters Gamma distributions (e.g., http://en.wikipedia.org/wiki/Gamma_distribution or http://www.itl.nist.gov/div898/handbook/eda/section3/eda366b.htm).
In principle this should be possible by means of the "scaling property": Gamma(shape,scale) ~ scale*Gamma(shape,1)
This would simply translate into boost::mt19937 rng; boost::gamma_distribution<> gamma(shape); // 1-parameter Gamma distribution boost::variate_generator
rvg(rng, gamma);
double r = scale*rvg();
However, when shape == 1 the implementation uses the fact that Gamma(1) ~ Exp(1). This is rigth when scale==1 but not when scale != 1 since the exact relation is Gamma(1,scale) ~ Exp(1/scale)
It is possibile to fix this problem? Also, it seems that the C++0x is rethinking the way to model the Gamma distribution as a two parameter distribution (see Sec. 26.5.8.3.3 [rand.dist.pois.gamma] at http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2009/n3000.pdf). Maybe it is the case that also Boost think about it, do you?
Please file a trac ticket.
Just done! See https://svn.boost.org/trac/boost/ticket/3888 Thanks!!! Best, -- Marco