Bill - I am attempting a Bayesian solution to the problem. As is often the case with this method, the hard part is coming up with a good prior probability. MCMC is another possibility I may look at. I looked at WinBUGS a long time ago - whether or not I re-visit it may depend on how other approaches go.
Thanks, Devon On 1/15/07, Bill Harris <[EMAIL PROTECTED]> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 "Devon McCormick" <[EMAIL PROTECTED]> writes: > thanks for your response. I guess I wasn't clear on what > I'm looking for: an approximation of a normal distribution > that has the same mean and standard deviation as the > actual distribution which will, in general, be non-normal. Devon, Is what you're trying to do what some might solve with Markov Chain Monte Carlo approaches and a hierarchical Bayesian approach using OpenBUGS (http://mathstat.helsinki.fi/openbugs/), WinBUGS (http://www.mrc-bsu.cam.ac.uk/bugs/welcome.shtml), or the like? Bill - -- Bill Harris http://facilitatedsystems.com/weblog/ Facilitated Systems Everett, WA 98208 USA http://facilitatedsystems.com/ phone: +1 425 337-5541 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.3 (MingW32) Comment: For more information, see http://www.gnupg.org iD8DBQFFrFue3J3HaQTDvd8RAnMdAJ0TWDdPZXz2E9RBfObP7/7vPRtCEgCeN9Kr 0ifH/0xIXfLixAv3RmPqBA4= =LdZY -----END PGP SIGNATURE----- ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
-- Devon McCormick ^me^ at acm. org is my preferred e-mail ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
