Re: [Boost-users] [distributions]: Inverse Gamma

Subject: Re: [Boost-users] [distributions]: Inverse Gamma
From: Thomas Mang (Thomas.Mang_at_[hidden])
Date: 2010-08-04 17:05:45

>>> (And there is the question of testing - some parameters and value combinations (preferably exact) are needed for
> sanity
>>> and accuracy checks).
>
> Any *exact* pdf and cdf values from *exact* parameters would be helpful - if you know any.  But I'll look more closely
> at the equations to try to find some too.

Sorry, no exact ones directly available.

>
> Do you have easy access to R that you could produce some test values (to double precision only of course)?

Yes I have, can produce pretty much whatever you would need.

R and C++ also have a language interface. If it helps I should be able 
to write a loop which evaluates some million or so calls of the 
functions using both R and the boost implementation and compares them.
If you just need some R evaluations output to a file, no problem as well.
Let me know what you would need.

best,
Thomas