The Interacting Behavior of Short- and Long-memory Investors in Agent-based Model

Tan, Duoduo (2007) The Interacting Behavior of Short- and Long-memory Investors in Agent-based Model. [Dissertation (University of Nottingham only)] (Unpublished)

[img] PDF - Registered users only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (881kB)

Abstract

Standard financial theories believe that the market price has included all information available and the past price cannot be used to forecast future price. However, in the real world, investors who use different lengths of past information to determine their investment are usually able to beat the market. The project tries to investigate whether the trading behavior of long- and short-memory investors could affect price process of the market through agent-based model. Although the technology of agent-based model is still imperfectly, it seems to provide a reasonable sense to the actual world.

The model is built on the base of Santa Fe artificial stock market and implemented by Java programming language. The Java implementation highly increases the transparency and approachability of the project. The modularity of the code makes extension and substitution of the composition part of the market easily.

Four main experiments are carried out and many interest results are obtained. Apart from the result that long-memory agents are able to dominate the market, the optimal market composition and the optimal investment horizons are derived.

Item Type: Dissertation (University of Nottingham only)
Depositing User: EP, Services
Date Deposited: 10 Mar 2008
Last Modified: 23 Oct 2016 00:45
URI: http://eprints.nottingham.ac.uk/id/eprint/21060

Actions (Archive Staff Only)

Edit View Edit View