Hi everyone...
I need desperately some serious help here...
and please forgive me for my English in advance.
I have a database of 4 quantities measured from a large number of
samples originating from different groups.
I'm giving the above figure..
sample(1) from group(1) quantity1 quantity2 quantity3 quantity4
...
sample(i) from group(1) q1 q2 q3 q4
sample(1) from group(2) q1 q2 q3 q4
...
sample(i) from group(2) q1 q2 q3 q4
sample(1) from group(i) q1 q2 q3 q4
...
sample(i) from group(i) q1 q2 q3 q4
hope this clears things up...
I'm trying to discriminate the above groups as much as I can using the 4
quantities above in a 2D plot.
I can take a pleasant discrimination by plotting a q1-q2 scatter plot,
but I have some overlaps that I would like to be eliminated.
Is there a way to (pre)set the groups and calculate the new x and y
(function of q1, q2, q3 and q4),
which maximize the discrimination of the groups in a scatter plot?
Please note that I'm not looking for the cluster analysis.
I already know which are the groups and I would like to plot them
together and discriminated.
Thanks in advance
Dimitris
_________________________________________________________________
Find what you need at prices you?ll love. Compare products and save at MSN®
Shopping.
http://shopping.msn.com/default/shp/?ptnrid=37,ptnrdata=24102&tcode=T001MSN20A0701
|