Driver Estimation in Non-Linear Autoregressive Models

Tom Dupré la Tour, Yves Grenier, Alexandre Gramfort

Abstract

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.
BibTeX
@inproceedings{tour2018driver,
  author = {Tom Dupré la Tour and Yves Grenier and Alexandre Gramfort},
  title = {{Driver Estimation in Non-Linear Autoregressive Models}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2018},
  doi = {10.1109/icassp.2018.8462268},
}
Figure