Driver Estimation in Non-Linear Autoregressive Models

Driver Estimation in Non-Linear Autoregressive Models

In non-linear autoregressive models, the time dependency of coefficients is often driven by a particular time-series which is not given and thus has to be estimated from the data. To allow model evaluation on a validation set, we describe a parametric approach for such driver estimation. After estim...

Full description

Saved in:
Bibliographic Details
Published in:2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) pp. 4519 - 4523
Main Authors: Duprela Tour, Tom, Grenier, Yves, Gramfort, Alexandre
Format: Conference Proceeding
Language:English
Published: IEEE 01-04-2018
Subjects:
Biological system modeling
Computational modeling
Couplings
cross-frequency coupling
Data models
electrophysiology
Estimation
non-linear autoregressive models
Optimization
spectrum estimation
Technological innovation
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In non-linear autoregressive models, the time dependency of coefficients is often driven by a particular time-series which is not given and thus has to be estimated from the data. To allow model evaluation on a validation set, we describe a parametric approach for such driver estimation. After estimating the driver as a weighted sum of potential drivers, we use it in a non-linear autoregressive model with a polynomial parametrization. Using gradient descent, we optimize the linear filter extracting the driver, outperforming a typical grid-search on predefined filters.
ISSN:2379-190X
DOI:10.1109/ICASSP.2018.8462268