F. James
Rohlf is interested in the applications of mathematical methods and statistics
(especially multivariate statistics) to problems in biology with emphasis
on morphometrics and the theory of systematics. Along with Robert Sokal,
he is the co-author of the popular text Biometry, now in its third
edition. My current research is concerned with the
development, evaluation, and application of new statistical methods for
use in geometric morphometrics. Over the last few years I have written
papers demonstrating how standard statistical methods such as principal
components analysis, canonical variates analysis, multiple regression,
and other methods can be adapted for the analysis of shape (by
approximating the non-Euclidean shape space by a tangent space
approximation). An important aspect of this new approach is that it is
possible to visualize the results of many types of statistical analyses
by generating shapes of hypothetical organisms. A current application involves
the visualization of estimates of ancestral species.
|