Sorry for cross-posting.

I need some help for the following problem:

I've got a number of objects. These objects are grouped. There are a 
number of independend variables for every object. But the dependend 
variables are only known for the groups.

Thus, I could identify the correlation using the mean values per group 
of the independet variables.

Is there any possibility to include more information of the objects 
(e.g. distribution, quartiles) into the correlation?

Thanks
Torsten Franz

.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to