Online Learning for Time Series Prediction
Machine Learning
2013-02-28 v1
Abstract
In this paper we address the problem of predicting a time series using the ARMA (autoregressive moving average) model, under minimal assumptions on the noise terms. Using regret minimization techniques, we develop effective online learning algorithms for the prediction problem, without assuming that the noise terms are Gaussian, identically distributed or even independent. Furthermore, we show that our algorithm's performances asymptotically approaches the performance of the best ARMA model in hindsight.
Cite
@article{arxiv.1302.6927,
title = {Online Learning for Time Series Prediction},
author = {Oren Anava and Elad Hazan and Shie Mannor and Ohad Shamir},
journal= {arXiv preprint arXiv:1302.6927},
year = {2013}
}
Comments
17 pages, 6 figures