In my humble opinion, one of the most important steps in learning MV statistics is acquiring an appreciation of the geometric aspects of the methods. Algebraic derivations are important, but an intuitive sense of rotations, projections, etc., is a major step in understanding what is being done. This becomes self-evident after a while. To that end, I find the following an excellent resource specifically for this kind of information:
Mathematical Tools for Applied Multivariate Analysis J. Douglas Carroll, Paul E. Green, with contributions by Anil Chaturvedi 1997. Academic Press -ds PS: I must acknowledge that many of my opinions, such as this one, were acquired many years ago in Jim Rohlf's MV class, which I have attended at every opportunity. Any foolishness, of course, is my own. On Tue, 2003-04-01 at 07:45, [EMAIL PROTECTED] wrote: > Hello, > I just came across this service by accident, but I am pleased that I > have. My name is Marc Laflamme, and I am presently undertaking a thesis > in multivariate morphometrics at Queen's University in Kingston, Canada. > With this said, I more or less stumbled into morphometrics by accident, > since my original project was purely taxonomic, but when it became > obvious that my fossils were not going to be classified by simple > taxonomy, I became consumed by a morphometrics project. So for the last > few months, I have been struggling through the matrix and algebraic > jargon in the hopes of finally understanding what my computer was doing > when performing a PCA! To make a long story short, I am overwhelmed by > many of the supposed "introductory" textbooks and papers which assume I > have 10 years of statistical background under my belt. Please don't > misunderstand me, I truly love the power associated with multivariate > statistics, and the types of problems that can only be solved through > morphometrics (to the point where I hope to upgrade my present project > to a Phd dissertation), but I would really appreciate any comments on > good morphometric papers/books/websites that i could read in order to be > able to tackle the harder literature in due time. > > I appreciate any comments or suggestions. > > Thank you, > Marc Laflamme > > > > _________________________________________________________________ > The new MSN 8: advanced junk mail protection and 2 months FREE* > http://join.msn.com/?page=features/junkmail > > == > Replies will be sent to list. > For more information see http://life.bio.sunysb.edu/morph/morphmet.html. -- Dennis E. Slice, Ph.D. Department of Biomedical Engineering Division of Radiologic Sciences Wake Forest University School of Medicine Winston-Salem, North Carolina, USA 27157-1022 Phone: 336-716-5384 Fax: 336-716-2870 == Replies will be sent to list. For more information see http://life.bio.sunysb.edu/morph/morphmet.html.
