Forecasting market prices in a supply chain game

  • Authors:

  • Christopher Kiekintveld;Jason Miller;Patrick R. Jordan;Lee F. Callender;Michael P. Wellman

  • Affiliations:

  • Computer Science and Engineering, University of Michigan, 2260 Hayward Avenue, Ann Arbor, MI 48109-2121, USA;Department of Mathematics, Stanford University, Building 380, Stanford, CA 94305, USA;Computer Science and Engineering, University of Michigan, 2260 Hayward Avenue, Ann Arbor, MI 48109-2121, USA;Computer Science and Engineering, University of Michigan, 2260 Hayward Avenue, Ann Arbor, MI 48109-2121, USA;Computer Science and Engineering, University of Michigan, 2260 Hayward Avenue, Ann Arbor, MI 48109-2121, USA

  • Venue:
  • Electronic Commerce Research and Applications
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Predicting the uncertain and dynamic future of market conditions on the supply chain, as reflected in prices, is an essential component of effective operational decision-making. We present and evaluate methods used by our agent, Deep Maize, to forecast market prices in the trading agent competition supply chain management game (TAC/SCM). We employ a variety of machine learning and representational techniques to exploit as many types of information as possible, integrating well-known methods in novel ways. We evaluate these techniques through controlled experiments as well as performance in both the main TAC/SCM tournament and supplementary Prediction Challenge. Our prediction methods demonstrate strong performance in controlled experiments and achieved the best overall score in the Prediction Challenge.