Abstract Details
Activity Number:
|
189
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 4, 2014 : 10:30 AM to 12:20 PM
|
Sponsor:
|
IMS
|
Abstract #313462
|
|
Title:
|
Covariance and Correlation for Multivariate Functional Data
|
Author(s):
|
Alexander Petersen *+ and Hans-Georg Müller
|
Companies:
|
University of California, Davis and University of California, Davis
|
Keywords:
|
Functional Data ;
Covariance ;
Partial Correlation
|
Abstract:
|
Assume one has $p$-dimensional functional data for $n$ subjects, $\{X_{i1}, X_{i2},\ldots,X_{ip}\}$, $i = 1, \ldots n$ on a common domain. The design setting for these data is flexible, allowing even for sparse longitudinal data. Given the utility and influence of a covariance matrix in standard multivariate analysis, we define an overall $p\times p$ functional covariance matrix $\Sigma$ by first transforming the pointwise covariance matrices, integrating, and transforming back. Such transformations may be based on specialized metrics for the nonlinear space of covariance matrices. This yields a Fr\'echet-like mean matrix, where the $(k,l)$-th element is a scalar measurement of covariance between processes $X_k$ and $X_l$. This is then extended to a functional partial correlation matrix. Estimates are obtained through standard cross-covariance estimates for paired functional data and consistency results then depend on the observational design setting. Examples of application areas include dimension reduction through eigenanalysis of $\Sigma$ and the construction of graphical models via partial correlations.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.