R-max - a general polynomial time algorithm for near-optimal reinforcement learning
The Journal of Machine Learning Research
JCAT: a platform for the TAC market design competition
Proceedings of the 7th international joint conference on Autonomous agents and multiagent systems: demo papers
IJCAI'99 Proceedings of the 16th international joint conference on Artifical intelligence - Volume 1
Censored exploration and the dark pool problem
Communications of the ACM
What the 2007 TAC Market Design Game tells us about effective auction mechanisms
Autonomous Agents and Multi-Agent Systems
A game-theoretic analysis of market selection strategies for competing double auction marketplaces
Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems: volume 1 - Volume 1
| Hi-index | 0.01 |
Traders that operate in markets with multiple competing marketplaces must often choose with which marketplace they will trade. These choices encourage marketplaces to seek competitive advantages against each other by adjusting various parameters, such as the price they charge, or how they match buyers and sellers. Traders can take advantage of this competition to improve utility. However, appropriate strategies must be used to decide with which marketplace a trader should shout. In this paper, we assess several different solutions to the problem of marketplace selection by running simulations of double auctions using the JCAT platform. The parameter spaces of these strategies are explored to find the best performing strategies. Results indicate that the softmax strategy is the most successful at maximising trader profit and global allocative efficiency in both adaptive and non-adaptive markets. The @e-decreasing strategy performs well in adaptive markets, while also showing greater stability in its parameter space than softmax. All marketplace selection strategies outperform the random marketplace selection strategy.