Model identification for ARMA time series through convolutional neural networks
Methodology
2020-07-21 v2 Machine Learning
Computation
Abstract
In this paper, we use convolutional neural networks to address the problem of model identification for autoregressive moving average time series models. We compare the performance of several neural network architectures, trained on simulated time series, with likelihood based methods, in particular the Akaike and Bayesian information criteria. We find that our neural networks can significantly outperform these likelihood based methods in terms of accuracy and, by orders of magnitude, in terms of speed.
Cite
@article{arxiv.1804.04299,
title = {Model identification for ARMA time series through convolutional neural networks},
author = {Wai Hoh Tang and Adrian Röllin},
journal= {arXiv preprint arXiv:1804.04299},
year = {2020}
}
Comments
17 pages, 1 figure, 4 tables