A Geometric Approach to Confidence Regions and Bands for Functional Parameters

Speaker: Matthew Reimher
Affiliation: Department of Statistics at Penn State University.
When: Friday 17th June 2016
Time: 15:00:00
Abstract: Functional data analysis, FDA, is now a well established discipline of statistics, with its core concepts and perspectives in place. Despite this, there are still fundamental statistical questions which have received relatively little attention. One of these is the systematic development of techniques for constructing confidence regions for functional parameters. I will present new work that takes a geometric approach to developing, understanding, and visualizing such regions. Simulations and an application to Fractional Anisotropy will also be presented. contact: