Hi,
I am trying to decide how best to approach a
regression in which the DV and IVs are all distance
matrices. QAP seems to be the standard approach in
work on social networks. However, my matrices are
3-dimensional, and I haven't seen an extension of QAP
to this case. I'm wondering if vectorizing the
matrices, then and using -permute- and -regress-
together will yield the same result: more accurate
standard errors (as I understand it, the coefficients
themselves are not problematic). Any thoughts on this
approach? Has anyone seen a comparison of the results
of QAP vs. premuation of a vectorized matrix?
Thanks
Zachary
-----------------
Zachary Neal, ABD
Managing Editor, City & Community
University of Illinois at Chicago
[email protected]
____________________________________________________________________________________
Check out the hottest 2008 models today at Yahoo! Autos.
http://autos.yahoo.com/new_cars.html
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
