s-news
[Top] [All Lists]

Multivariate Normal Samples with Correlation among Rows

To: S-News <s-news@lists.biostat.wustl.edu>
Subject: Multivariate Normal Samples with Correlation among Rows
From: Paul Lasky <phlasky@earthlink.net>
Date: Mon, 21 Aug 2006 13:27:52 -0700
Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=dk20050327; d=earthlink.net; b=ipjNL7ALtRWVmdjz5XVapmSuaMGEvjREhlf1kvv7+gYsZd8o1IuYB4PFOswCUoww; h=Message-ID:Date:From:Reply-To:User-Agent:To:Subject:Mime-Version:Content-Transfer-Encoding:Content-Type:X-ELNK-Trace:X-Originating-IP;
Reply-to: phlasky@earthlink.net
User-agent: Thunderbird 1.5.0.5 (Windows/20060719)
I have a n X m matrix M that is positive definite ( positive det of the associated covar matrix). The columns index subjects and rows are index time. I wish to random sample from the matrix assuming a multivariate normal model, BUT I wish to take into account the time correlation, i.e. cor( t ( M ) ).

How do I do this, or is it even possible ? I know that ns random vectors are available from a standard-type multivariate normal model by rmvnorm( ns, colMeans( M ), var( M ) )

 Paul H. Lasky
 P & B Consultants




--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.11.3/423 - Release Date: 8/18/2006


<Prev in Thread] Current Thread [Next in Thread>
  • Multivariate Normal Samples with Correlation among Rows, Paul Lasky <=