UNIGE document Chapitre d'actes
previous document  unige:103203  next document
add to browser collection
Title

Learning Predictive Leading Indicators for Forecasting Time Series Systems with Unknown Clusters of Forecast Tasks

Authors
Published in Proceedings of the Ninth Asian Conference on Machine Learning (ACML 2017). Seoul (Korea) - 15-17 November 2017 - . 2017, p. 161-176
Abstract We present a new method for forecasting systems of multiple interrelated time series. The method learns the forecast models together with discovering leading indicators from within the system that serve as good predictors improving the forecast accuracy and a cluster structure of the predictive tasks around these. The method is based on the classical linear vector autoregressive model (VAR) and links the discovery of the leading indicators to inferring sparse graphs of Granger causality. We formulate a new constrained optimisation problem to promote the desired sparse structures across the models and the sharing of information amongst the learning tasks in a multi-task manner. We propose an algorithm for solving the problem and document on a battery of synthetic and real-data experiments the advantages of our new method over baseline VAR models as well as the state-of-the-art sparse VAR learning methods.
Keywords Time series forecastingVARGranger causalityStructured sparsityMultitask learningLeading indicators
Full text
Structures
Research groups Computer Vision and Multimedia Laboratory
Viper group
Geneva Artificial Intelligence Laboratory
Citation
(ISO format)
GREGOROVA, Magda, KALOUSIS, Alexandros, MARCHAND-MAILLET, Stéphane. Learning Predictive Leading Indicators for Forecasting Time Series Systems with Unknown Clusters of Forecast Tasks. In: Proceedings of the Ninth Asian Conference on Machine Learning (ACML 2017). Seoul (Korea). [s.l.] : [s.n.], 2017. p. 161-176. https://archive-ouverte.unige.ch/unige:103203

65 hits

14 downloads

Update

Deposited on : 2018-03-26

Export document
Format :
Citation style :