Scalability and Optimisation of a Committee of Agents Using Genetic Algorithm - Computer Science > Multiagent SystemsReport as inadecuate




Scalability and Optimisation of a Committee of Agents Using Genetic Algorithm - Computer Science > Multiagent Systems - Download this document for free, or read online. Document in PDF available to download.

Abstract: A population of committees of agents that learn by using neural networks isimplemented to simulate the stock market. Each committee of agents, which isregarded as a player in a game, is optimised by continually adapting thearchitecture of the agents using genetic algorithms. The committees of agentsbuy and sell stocks by following this procedure: (1) obtain the current priceof stocks; (2) predict the future price of stocks; (3) and for a given pricetrade until all the players are mutually satisfied. The trading of stocks isconducted by following these rules: (1) if a player expects an increase inprice then it tries to buy the stock; (2) else if it expects a drop in theprice, it sells the stock; (3)and the order in which a player participates inthe game is random. The proposed procedure is implemented to simulate tradingof three stocks, namely, the Dow Jones, the Nasdaq and the SandP 500. A linearrelationship between the number of players and agents versus the computationaltime to run the complete simulation is observed. It is also found that noplayer has a monopolistic advantage.



Author: T. Marwala, P. De Wilde, L. Correia, P. Mariano, R. Ribeiro, V. Abramov, N. Szirbik, J.Goossenaerts

Source: https://arxiv.org/







Related documents