From: Przemyslaw Sliwa <sliwa@euv-frankfurt-o.de>
To: <gsl-discuss@sources.redhat.com>
Subject: Re: multivariate gaussian distribution (Code)
Date: Mon, 29 Dec 2003 10:41:00 -0000 [thread overview]
Message-ID: <47804.160.83.32.14.1072694506.squirrel@webmail.euv-frankfurt-o.de> (raw)
Dear All,
There are several thing I cannot understand in the code. I suppose it has
been designed to handle real random vectors for multivariate normal
distribution. Therefore at least two points:
1. Why did you (Emmanuel) use the Eigenvalue decomposition of the
covariance matrix. This method is extremely inefficient with the gsl
eigenvalues code. One shall (I do) use the Cholesky decomposition of the
covariance matrix in order to compute the lower triangular matrix L of the
form Cov = LL' This procedure is described in several books e.g.
Harville's "Matrix Algebra from a Statistician's perspective".
2. Why do you use this Box Mueller Algorithm? My point is: if you have a
vector X of independent, normally distributed variables (Covariance matrix
equals identity matrix) the product P = LX is always normally distributed
with covariance matrix Cov, since (assuming E(X) = 0) E(PP') = E(LX(LX)')
= E(LXX'L) = LIL' = Cov. Similar to this method one can simulate the whole
family of elliptically countered distributions (like Bessel, generalized
Lapalce, t-distributions).
Did I miss something?
Thanks for explanations,
Kind regards,
Przem
next reply other threads:[~2003-12-29 10:41 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-12-29 10:41 Przemyslaw Sliwa [this message]
-- strict thread matches above, loose matches on Subject: below --
2003-12-30 9:47 Emmanuel Benazera
2003-12-30 10:55 ` Przemyslaw Sliwa
2003-12-30 11:31 ` Przemyslaw Sliwa
2003-12-29 23:02 Przemyslaw Sliwa
2003-12-29 17:54 Emmanuel Benazera
2003-12-29 11:52 Przemyslaw Sliwa
2003-12-20 11:10 multivariate gaussian distribution Emmanuel Benazera
2003-12-22 14:21 ` Brian Gough
2003-12-28 10:29 ` multivariate gaussian distribution (Code) Emmanuel Benazera
2003-12-29 11:41 ` Brian Gough
2003-12-29 16:47 ` Emmanuel Benazera
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=47804.160.83.32.14.1072694506.squirrel@webmail.euv-frankfurt-o.de \
--to=sliwa@euv-frankfurt-o.de \
--cc=gsl-discuss@sources.redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).