Rice University logoGeorge R. Brown School of Engineering
 
Statistics
 

Theofanis Sapatinas
Professor of Statistics
University of Cyprus

Bootstrap-Based Testing of Equality of Mean Functions or Equality of Covariance Operators for Functional Data

We investigate properties of a simple and easy to implement bootstrap  methodology for testing equality of mean functions or equality of covariance operators  between different  populations of functional data. The methodology bootstraps the original functional dataset in  a way that satisfies  the null hypothesis of interest. Theoretical size and power results are derived  for some interesting test statistics, whose limiting distributions depend on unknown and difficult-to-estimate infinite-dimensional parameters.  Simulations demonstrate good size and power performances of the bootstrap-based functional tests. An illustration  to an interesting real-life dataset is also presented. (This is joint work with Efstathios Paparoditis, University of Cyprus, Cyprus.)

Monday, February 6, 2017
4:00 PM to 5:00 PM
Duncan Hall, RM1070