[19392] in s-news-athena
[S] joint normalization
daemon@ATHENA.MIT.EDU (Rosenfeld, Simon (NCI))
Mon Oct 4 12:48:18 1999
Message-Id: <33A032D3B953D2119D750008C7A48E68024D2AD9@nihexchange7.nih.gov>
From: "Rosenfeld, Simon (NCI)" <rosenfes@mail.nih.gov>
To: s-news@wubios.wustl.edu
Date: Mon, 4 Oct 1999 12:35:45 -0400
Mime-Version: 1.0
Content-Type: text/plain
Dear colleagues:
I have m variables with N observations of each of them.
All the variables are distinctly non-normal and have different PDFs,
parameters of scale and location. I need to find a SINGLE non-linear
transformation
for all the variables in such a way as to make all the variables (after
standardization)
1)as close to multi-dimensional normal as possible
2)as close to MUTUALLY linear as possible
3)as close to multiple-linear as possible
The questions:
1)Are there any S-plus functions performing this task
2)Any literature on the subject
3)Any experience in doing this kind of things
Thank you.
Simon Rosenfeld,
NIH/NCI
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
