Dear members:
I am wondering
(1) if there is genetic statistics or genetic epidemiology discussion group
exist;
(2) I am going to analyse data from twins observed in two time periods (and
at the baseline), to study whether there is genetic contribution to a
continuous trait response to the
If it makes sense to represent your data as 4 column vectors, say y1 and y2
for the first set and x1 and x2 for the second, then you may consider to
calculate the canonical correlation between (y1,y2) and (x1,x2).
Jos Jansen
"Peter J. Wahle" [EMAIL PROTECTED] wrote in message
WcQA6.745$[EMAIL
Your best bet for a newsgroup is: bionet.molbio.gene-linkage. While the
title has "linkage" in it, the people who read reply to it are usually
well-versed in a variety of different statistical genetic approaches to
analysis of genetic data. The signal:noise ratio is very low (as are
total
In article [EMAIL PROTECTED], Jay Warner [EMAIL PROTECTED] wrote:
one tech issue, one thinking issue, I believe.
1) Tech: if np _and_ n(1-p) are 5, the distribution of binomial
observations is considered 'close enough' to Normal. So 'large n' is
OK, but fails when p, the p(event), gets
