Mail Archive Search: <+subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*beta\s+function\,numerical\s+methods\,amoeba\s*$/>

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*beta\s+function\,numerical\s+methods\,amoeba\s*$/: 4 ]

Total 4 documents matching your query.

1. beta function,numerical methods,amoeba (score: 1): Author: "Steve Su" <s.su@qut.edu.au>; Date: Wed, 19 Feb 2003 17:40:07 +1000; Dear All, I am having great difficulties with using numerical methods to solve equations that involve beta function. For illustration purposes I will show only the beta function part of the equation.; /archives/html/s-news/2003-02/msg00119.html (13,682 bytes)
2. Re: beta function,numerical methods,amoeba (score: 1): Author: "Liaw, Andy" <andy_liaw@merck.com>; Date: Wed, 19 Feb 2003 09:19:56 -0500; See if this helps (optim() is in the most recent version of MASS): + exp( lgamma(a) + lgamma(b) - lgamma(a+b) ) + } + x<-coef[1] + y<-coef[2] + sqrt((beta(x,y)-0.5)^2) + } $par: [1] 3.1686527 0.77628; /archives/html/s-news/2003-02/msg00122.html (18,736 bytes)
3. Re: beta function,numerical methods,amoeba (score: 1): Author: "Liaw, Andy" <andy_liaw@merck.com>; Date: Wed, 19 Feb 2003 11:46:19 -0500; [cc to s-news as I supposed is intended.] Besides, why square beta(x, y) - 0.5 and then take square root? Why not just use abs()? Andy -- Notice: This e-mail message, together with any attachments, c; /archives/html/s-news/2003-02/msg00123.html (24,714 bytes)
4. Re: beta function,numerical methods,amoeba (score: 1): Author: Spencer Graves <spencer.graves@PDF.COM>; Date: Wed, 19 Feb 2003 08:52:49 -0800; If you use square the deviation rather than using absolute value, you might get better numerical performance, e.g., from an algorithm that assumes the function is differentiable at the optimum! Spenc; /archives/html/s-news/2003-02/msg00124.html (22,276 bytes)

Current List: 1 - 4
Page: [1]