[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
more on gamma
From: |
Michael Creel |
Subject: |
more on gamma |
Date: |
Wed, 26 Mar 2003 11:47:03 -0800 |
User-agent: |
KMail/1.5.1 |
Thanks to a couple of you for suggestions. Two solutions to the gamma overflow
problem are:
1) to write a wrapper function, as follows:
function [f] = mc_gamma(z)
test = z > 170;
percentage = mean(test);
if sum(test) > 0
warning("mc_gamma: %f of the arguments were bounded",
percentage);
endif
z = 170*test + (1-test).*z;
f = gamma(z);
endfunction
This prevents overflows and gives you a warning when it occurs.
2) Run the input (x) to gamma(g(x)) through a nonlinear map that bounds it so
that eps < g(x) < 170, e.g,
alpha = 165/(1 + exp(5 - psi(k+1,:))) + eps; # eps < alpha < 165
Now gamma(alpha) won't crash.
These two solutions together seem to be working for me, and were easy to
implement.
Thanks, Michael
-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.
Octave's home on the web: http://www.octave.org
How to fund new projects: http://www.octave.org/funding.html
Subscription information: http://www.octave.org/archive.html
-------------------------------------------------------------
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- more on gamma,
Michael Creel <=