What is Geometric Statistics?

Speaker: Carlos Soto (Umass)

In this talk I'll briefly cover three large topics in geometric statistics; information geometry, shape analysis, and statistics on manifolds. In information geometry we consider probability measures while in shape analysis we consider (among other things) space curves yet these two areas have many commonalities. Both areas embed the objects of interest onto a Riemannian manifold. In statistics on manifolds one rather considers probability measures on manifolds. That is, how are ideas such as mean, covariance, principal component analysis, etc. etc. extended into non-Euclidean space. Time permitting, I'll mention some ongoing work on Lie groups.

 Date and location: 02/18 10:00 a.m. to 11:00 a.m. in LGRT 1685