Please use this identifier to cite or link to this item: http://dx.doi.org/10.14279/depositonce-12238
For citation please use:
Main Title: Adaptive Online Learning for the Autoregressive Integrated Moving Average Models
Author(s): Shao, Weijia
Radke, Lukas Friedemann
Sivrikaya, Fikret
Albayrak, Sahin
Type: Article
URI: https://depositonce.tu-berlin.de/handle/11303/13452
http://dx.doi.org/10.14279/depositonce-12238
License: https://creativecommons.org/licenses/by/4.0/
Abstract: This paper addresses the problem of predicting time series data using the autoregressive integrated moving average (ARIMA) model in an online manner. Existing algorithms require model selection, which is time consuming and unsuitable for the setting of online learning. Using adaptive online learning techniques, we develop algorithms for fitting ARIMA models without hyperparameters. The regret analysis and experiments on both synthetic and real-world datasets show that the performance of the proposed algorithms can be guaranteed in both theory and practice.
Subject(s): ARIMA model
time series analysis
online optimization
online model selection
Issue Date: 29-Jun-2021
Date Available: 26-Jul-2021
Language Code: en
DDC Class: 000 Informatik, Informationswissenschaft, allgemeine Werke
Sponsor/Funder: DFG, 414044773, Open Access Publizieren 2021 - 2022 / Technische Universität Berlin
Journal Title: Mathematics
Publisher: MDPI
Volume: 9
Issue: 13
Article Number: 1523
Publisher DOI: 10.3390/math9131523
EISSN: 2227-7390
TU Affiliation(s): Fak. 4 Elektrotechnik und Informatik » Inst. Wirtschaftsinformatik und Quantitative Methoden » FG Agententechnologien in betrieblichen Anwendungen und der Telekommunikation (AOT)
Appears in Collections:Technische Universität Berlin » Publications

Files in This Item:
mathematics-09-01523-v2.pdf
Format: Adobe PDF | Size: 1.78 MB
DownloadShow Preview
Thumbnail

Item Export Bar

This item is licensed under a Creative Commons License Creative Commons