|
Activity Number:
|
345
- Time Series and High-Dimensional Data
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
Business and Economic Statistics Section
|
|
Abstract #313298
|
|
|
Title:
|
Principal Component Analysis Using Frequency Components of Multivariate Time Series
|
|
Author(s):
|
Raanju Sundararajan*
|
|
Companies:
|
Southern Methodist University
|
|
Keywords:
|
multivariate time series;
spectral matrix;
principal component analysis;
forecasting
|
|
Abstract:
|
Dimension reduction techniques for multivariate time series decompose the observed series into a few useful independent/orthogonal univariate components. We develop a spectral domain method for multivariate second-order stationary time series that linearly transforms the observed series into a few groups of lower-dimensional multivariate subseries. These multivariate subseries have non-zero spectral coherence among components within a group but have zero spectral coherence among components across groups. The observed series is expressed as a sum of frequency components whose variances are proportional to the spectral matrices at the respective frequencies. The transformation matrix is then estimated using an eigendecomposition of the sum of variance matrices of these frequency components its asymptotic properties are derived. Finally, a consistent test on the cross-spectrum of pairs of components is used to find the desired segmentation into the lower-dimensional subseries. The numerical performance of the proposed method is illustrated through simulation and real data examples.
|
Authors who are presenting talks have a * after their name.
