Research And Implementation Of Futures Risk Aversion System Based On Association Rules

With the rapid development of economy, the futures trading plays an more and more important role in the world economy. As a kind of advanced commodity transaction form and a modern investment method, the futures has a huge development potential. Due to the flexibility and a strong leverage effect of the futures market, the futures trading is not only full of opportunities, but also full of risks. To the investors, it is very important to grasp the opportunity and avoid risks. And to the agents of investors, namely the futures commission merchants, it is also very important. As an important tool for knowledge discovery, data mining can find the valuable information from a large amount of data. And this is just appropriate for the large number of transaction records of futures trading. As a financial data, these records is of high integrity, and also contain all sorts of trading information of traders. In order to study the effect of the trading behaviors and habits of traders on the futures risk, we use association rules to analyse the transaction records of futures traders, to find out the potential relation between trading activity and risk, and find the related rules, and achieve the risk early warning and risk aversion. This paper studies the futures risk aversion system based on the association rules, the main work is as follows:(1)This paper introduces the relevant background knowledge of futures trading, discusses the tradind features, the market mechanism and the development at home and abroad. The basic theory and basic method of association rules are introduced. And the process of association rule mining is illustrated in this paper. According to the differences of rules, the association rules are classified. And the Apriori algorithm is analysed, introduces the algorithm thought, content, nature, detailing the specific steps of the algorithm and present the pseudo code description.(2)Studies and implements a futures risk aversion system, analyses the system demand and function design. According to the risk traders faced in the futures market, combine the association rule mining and risk aversion, and calculate the risk behavior rules using association mining, help the futures companies to achieve early warning of risks and realize the risk aversion.(3)Proposed an improved algorithm based on matrix processing of Apriori. The improved algorithm uses a Boolean matrix to store the transactions in the transaction database. Through the use of the nature of the Apriori and inference to operating the matrix to search the frequent item sets, it avoids the repeated process of scanning the entire transaction library, improve the efficiency of the algorithm.(4) According to the relationship between futures risk and the traders' behavior in the futures market, the risk behavior rule is put forward, expressed by association rules.After the implement, the system is tested with the real future trading data, and the risk behavior rule can be calculated, which reach the expected effect.