Detection Technology Research Of The Futures Market Manipulation Behavior

In the futures market, market manipulation brings great harm to the development of financial markets. That is the reason why market manipulation has always been the focus of regulatory agencies. Timely detecting and finding the market manipulation from a large number of normal trading behavior is a prerequisite for prevention and supervision. In view of the current electronic trading market-oriented situation, how to effectively use computer technology detect the market manipulation from a batch of transactions has become one of the main problems of financial market supervision.Currently, detecting the illegal market manipulation is mainly depending on personal experience and data statistics. However, with the growing volume of futures’ market trading and the increasing complexity of market manipulation and higher accuracy and timeliness is required by regulatory agencies, the difficulty of finding such behavior is rising exponentially.Considering the main indicators on market manipulation detection(commission time, the amount of commission, commission varieties), using cluster analysis technology and based on an actual trading data of futures markets, an algorithm which applied to large-scale transaction data is proposed to provide a viable technique for futures trading market to find the illegal operating behavior.The main findings are as follows:(1) Analysis the current status of futures market manipulation, the main flow and the key technologies of detecting operation. Pointing out the insufficient aspect of many existed methods which directly applied to futures market manipulation. Elaborating the definition, the forms, the patterns and the analyze processes of market manipulation. And introducing the clustering analysis methods which applied to market manipulation detection, these technologies will provide a theoretical basis and technical support for the market manipulation detection.(2)The core issue of futures markets is how to quickly find the manipulation behavior under the limited time and space. In order to solve that problem, an algorithm named Hierarchical Clustering Algorithm based on Orthogonal List is proposed. This method treats orthogonal list as a storage structure. The transactions similarity matrix is mapped into the cross chain, so the traversal of the database record converted to traverse the cross-linked list. Then, the transaction behaviors with high similarity will be put in the same cluster according to hierarchical clustering method. And the clusters with relative higher similarity will be listed in the suspicious group to monitor for future reference card.(3) The Futures Market Operating Behavior Detection System will be achieved based on the above studies. First, proposing the system design; Second, the key modules will be detailed design according to the system design. Finally, The Futures Market Operating Behavior Detection System is implemented based on the above work. The effectiveness of the algorithm and the system is verified by the futures exchange’s business data which has applied to the system. The results show that the proposed method which can greatly compress the amount of data is suitable for detecting manipulation of large-scale transactions. It can improve the timeliness and accuracy of detection, and provide effective anti-manipulation regulation technology for futures market.