Kernel Methods for Nonlinear Connectivity Detection
Signal Processing
2019-07-24 v1 Dynamical Systems
Applications
Other Statistics
Abstract
In this paper, we show that the presence of nonlinear coupling between time series may be detected employing kernel feature space representations alone dispensing with the need to go back to solve the pre-image problem to gauge model adequacy. As a consequence, the canonical methodology for model construction, diagnostics, and Granger connectivity inference applies with no change other than computation using kernels in lieu of second-order moments.
Cite
@article{arxiv.1806.07440,
title = {Kernel Methods for Nonlinear Connectivity Detection},
author = {Lucas Massaroppe and Luiz A. Baccalá},
journal= {arXiv preprint arXiv:1806.07440},
year = {2019}
}
Comments
14 pages, 14 figures, preliminary version being prepared for submission to a refereed journal