		/vol06/ishibuchi-oh/	© Copyright 1998
		Volume 06	Received: Accepted:	01 Jul 1998 15 Oct 1998

Competition between strategies for a market selection game Hisao Ishibuchi, Chi-Hyon Oh, Tomoharu Nakashima
Abstract
	Our game in this paper is a non-cooperative repeated game Êwhere many players compete with one another at several markets. At each iteration of our game, each player is supposed to choose a single market Êfor maximizing his own profit obtained by selling his product at the selected market. It is assumed in our market selection game that the market price of the product is determined by the demand-supply relation at each market. For example, if many players bring their products to a particular market, the market price at that market becomes low. On the contrary, the market price is high if the total amount of products brought to the market is small. In this manner, the market price at each market is determined by the actions of all players. This means that the profit of each player depends on the actions of the other players. The main aim of this paper is to numerically analyze the competition between several strategies for our market selection game. We examine six strategies: a random selection strategy, a minimum transportation cost strategy, an optimal strategy for the previous actions, a mimic strategy of the nearest neighbor player, a Q-learning-based strategy, Êand a fuzzy ÊQ-learning-based strategy. The performance of each strategy is examined by computer simulations with 100 players and five markets. First we examine the performance of each strategy by assuming that all the players use the same strategy. Next we examine the competition between two strategies, each of which are used by half of the players. Then we examine the competition between two strategies further by changing the number of players adopting each strategy. Finally we perform computer simulations of the competition between various strategies where the number of players adopting each strategy is increased or decreased according to the performance of the strategy.

Full Text

Multimedia Links
(none)

Reference Links
(none)

Citation Reference

Get viewers
for PS & PDF

[CI Editor] [Site Manager]