Title: Multi-event decision making over multivariate time series

Authors: Chun-Kit Ngan; Alexander Brodsky; Jessica Lin

Addresses: Department of Computer Science, George Mason University, 4400 University Drive MSN 4A5, Fairfax, Virginia 22030-4422, USA ' Department of Computer Science, George Mason University, 4400 University Drive MSN 4A5, Fairfax, Virginia 22030-4422, USA ' Department of Computer Science, George Mason University, 4400 University Drive MSN 4A5, Fairfax, Virginia 22030-4422, USA

Abstract: We propose a multidimentional time-point model and algorithm to solve multi-event expert query parametric estimation (ME-EQPE) problems over multivariate time series. Our proposed model and algorithm combine the strengths of both domain-knowledge-based and formal-learning-based approaches to learn optimal decision parameters for maximising utility over multivariate time series. More specifically, our approach solves the decision optimisation problems to maximise the utility from multiple decision time points, as well as maintaining an optimality of the learned multiple sets of decision parameters in their respective events during the computations. We show that our approach produces a reasonable forecasting result by using the learned multiple sets of decision parameters.

Keywords: statistical learning; prediction; multivariate time series; decision support; multi-event decision making; modelling; expert query parametric estimation; decision optimisation; forecasting.

DOI: 10.1504/IJIDS.2013.055925

International Journal of Information and Decision Sciences, 2013 Vol.5 No.3, pp.263 - 282

Published online: 28 Feb 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article