[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: binocdf accuracy
|
From: |
Marco Caliari |
|
Subject: |
Re: binocdf accuracy |
|
Date: |
Mon, 08 Jul 2013 15:45:37 +0200 (CEST) |
|
User-agent: |
Alpine 1.00 (DEB 882 2007-12-20) |
Dear all,
I think the problem is not in betainc, but in the line
cdf(k) = 1 - betainc (p, tmp + 1, n - tmp);
in betaocdf (1-something_very_small). What about to replace this line with
cdf(k) = quadgk(@(t) t.^tmp.*(1-t).^(n-tmp-1),p,1)/beta(tmp+1,n-tmp)
?
If n or p are not scalar, unfortunately the previous line requires a loop.
Marco
- binocdf accuracy, Dr. Alexander Klein, 2013/07/06
- Re: binocdf accuracy,
Marco Caliari <=