摘要
连续双向拍卖市场中交易策略的设计问题远比单向拍卖复杂,本文首先检验了该市场中交易价格的马尔可夫性质,然后据此提出了基于马尔可夫链的自学习动态交易策略,最后通过比较实验发现,该策略明显优于"约束型零信息"策略。
The work to design trading strategy in a continuous double auction market is more complex than that in one--side auction. This paper firstly tests the Markov character of the sequence of transaction price in the continuous double auction market with ZI--C (Zero--Intelligence with Constraint) traders. Then a new trading strategy based on the Markov chain is provided. Finally, the new trading strategy is proved to be superior to the ZI--C strategy by control experiments.
出处
《中国管理科学》
CSSCI
2008年第1期111-116,共6页
Chinese Journal of Management Science
基金
国家自然科学基金资助项目(70301013)
关键词
连续双向拍卖
交易策略
马尔可夫链
continuous double auction
trading strategy
Markov chain