Statistics 5401
Multivariate Statistical Methods
Fall 2002
Please note: there is a mistake in the
printed course schedule (the web schedule is correct). The
course meets in Ford 115.
In this course we statistical methods that
handle more than one random variable at a time. We will
build some background on matrix operations and the multivariate
normal distribution. Then we'll move into multivariate analogs
of t-tests, ANOVA, and the like. After that we'll begin modeling
covariance structures, using principal components, factor analysis,
and canonical correlations. Finally, we'll finish up with some
work on classification (discriminant analysis) and clustering.
Many course materials are available in the
"Portable Document Format" (PDF), which can be viewed
and printed by using Acrobat Reader software. Versions of Acrobat
Reader for most computing platforms may obtained for free from
Adobe Inc.
G. W. Oehlert gary@stat.umn.edu
Last modified September 3, 2002.