Online Learning for Time Series Prediction

Oren Anava, Elad Hazan, Shie Mannor, Ohad Shamir
Proceedings of the 26th Annual Conference on Learning Theory, PMLR 30:172-184, 2013.

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, \emphwithout 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 this Paper


BibTeX
@InProceedings{pmlr-v30-Anava13, title = {Online Learning for Time Series Prediction}, author = {Anava, Oren and Hazan, Elad and Mannor, Shie and Shamir, Ohad}, booktitle = {Proceedings of the 26th Annual Conference on Learning Theory}, pages = {172--184}, year = {2013}, editor = {Shalev-Shwartz, Shai and Steinwart, Ingo}, volume = {30}, series = {Proceedings of Machine Learning Research}, address = {Princeton, NJ, USA}, month = {12--14 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v30/Anava13.pdf}, url = {https://proceedings.mlr.press/v30/Anava13.html}, 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, \emphwithout 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.} }
Endnote
%0 Conference Paper %T Online Learning for Time Series Prediction %A Oren Anava %A Elad Hazan %A Shie Mannor %A Ohad Shamir %B Proceedings of the 26th Annual Conference on Learning Theory %C Proceedings of Machine Learning Research %D 2013 %E Shai Shalev-Shwartz %E Ingo Steinwart %F pmlr-v30-Anava13 %I PMLR %P 172--184 %U https://proceedings.mlr.press/v30/Anava13.html %V 30 %X 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, \emphwithout 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.
RIS
TY - CPAPER TI - Online Learning for Time Series Prediction AU - Oren Anava AU - Elad Hazan AU - Shie Mannor AU - Ohad Shamir BT - Proceedings of the 26th Annual Conference on Learning Theory DA - 2013/06/13 ED - Shai Shalev-Shwartz ED - Ingo Steinwart ID - pmlr-v30-Anava13 PB - PMLR DP - Proceedings of Machine Learning Research VL - 30 SP - 172 EP - 184 L1 - http://proceedings.mlr.press/v30/Anava13.pdf UR - https://proceedings.mlr.press/v30/Anava13.html AB - 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, \emphwithout 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. ER -
APA
Anava, O., Hazan, E., Mannor, S. & Shamir, O.. (2013). Online Learning for Time Series Prediction. Proceedings of the 26th Annual Conference on Learning Theory, in Proceedings of Machine Learning Research 30:172-184 Available from https://proceedings.mlr.press/v30/Anava13.html.

Related Material