Online Program Home
  My Program

Abstract Details

Activity Number: 5 - New Developments in Functional Data
Type: Invited
Date/Time: Sunday, July 30, 2017 : 2:00 PM to 3:50 PM
Sponsor: IMS
Abstract #322233
Title: Clusering functional data using projections
Author(s): Aurore Delaigle* and Peter Hall and Tung Pham
Companies: University of Melbourne and University of Melbourne and University of Melbourne
Keywords: FDA ; clustering ; functional data

We show that, in the functional data context, by appropriately exploiting the functional nature of the data, it is possible to cluster the observations asymptotically perfectly. We demonstrate that this level of performance can often be achieved by the k-means algorithm as long as the data are projected on a carefully chosen finite dimensional space. We propose an iterative algorithm to choose the projection functions in a way that optimises clustering performance, where, to avoid peculiar solutions, we use a weighted least-squares criterion.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

Copyright © American Statistical Association