> I don't find in any statistics book the formula to evaluate the correlation > coefficient r for linear regression through origin. Please help me!
Thousands, or rather millions of other people more qualified to say this, and a similar number of ways to put/explain it, but the need for help apears to be urgent, so: there is none, because it's the same as in the usual case! - Linear transformations (like centering) don't affect r.
Gaj Vidmar University of Ljubljana, Faculty of Medicine, Institute of Biomedical Informatics