Skip to main navigation Skip to search Skip to main content

Stochastic beliefs learning and commodity futures price dynamics

  • Tan Li*
  • , Zhong Ying Qi
  • , Hong Yuan Niu
  • *Corresponding author for this work
  • School of Management, Harbin Institute of Technology
  • Small and Medium Enterprise Credit Department

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We propose a multi-agent-based artificial futures market to understand commodity futures price dynamics, which is called stylized facts. The multi-agent-based artificial futures market includes hedger agents and speculative agents. We use Brenner's stochastic belief learning model (SBL) to describe speculative agents' learning process. The SBL is a social learning process with local information. New market price is generated though a sealed-bid auction clearance mechanism. Our simulation result can reproduce the important observed stylized facts of futures price. The simulation programming language is Matlab 7. Our results show clustered volatility depends on speculator agents' imitations which are caused by social learning process. Using this simulation model, we can find futures price volatility has close relation with large speculator agents' trading activity.

Original languageEnglish
Title of host publication2010 International Conference on Management Science and Engineering, ICMSE 2010
Pages1186-1191
Number of pages6
DOIs
StatePublished - 2010
Externally publishedYes
Event17th International Conference on Management Science and Engineering, ICMSE 2010 - Melbourne, VIC, Australia
Duration: 24 Nov 201026 Nov 2010

Publication series

Name2010 International Conference on Management Science and Engineering, ICMSE 2010

Conference

Conference17th International Conference on Management Science and Engineering, ICMSE 2010
Country/TerritoryAustralia
CityMelbourne, VIC
Period24/11/1026/11/10

Keywords

  • Clustered volatility
  • Futures price
  • Heterogeneous agent
  • Stochastic belief learning

Fingerprint

Dive into the research topics of 'Stochastic beliefs learning and commodity futures price dynamics'. Together they form a unique fingerprint.

Cite this