Study On Intelligent Method Applications Of Financial Market Spillover And Anti-Money Laundering

Our daily life needs financial field, we make contract with money and financial market. And it is very significant to study and analyze financial market. With the development of information technology, plenty of data has been accumulated till today, with kinds of format. Analyzing and processing these data is very important in practices, helping us obtain valuable technology from "the information ocean". This paper investigates the complicated relationship in financial market with the technology of artificial intelligence, data mining and social network analysis et al. From the macroscopical perspective, we study the spillover effects between different international financial markets; from the microcosmic perspective, we investigate some problems in the anti-money laundering application, which can help commercial bank detect and predict the anti-money laundering.In summary, the thesis includes the following works:(1) Study the inter-linkages between the US and Shanghai and the US and HK stock market returns by wavelet multi-resolution analysis. The stock daily return time series signal was decomposed on different frequency bands to study the correlativity, and the energy proportions of different frequency components to the original signal were compared. The results show that the high frequency detail components represent much more energy than the low frequency smooth components, and that the movements in stock returns are mainly caused by the short term factors. Interdependent analysis between the high frequency detail components shows that the volatility spillover effects exist from the US to HK stock market, but does not exist from the US to Shanghai stock market, and the Shanghai stock market seems to be independence.(2) Investigate the information spillover effects between financial markets in the economic community by wavelet multi-resolution analysis. In this paper, wavelet multi-resolution decomposition is used to study the spillover effects of copper future returns between the two markets. The daily return time series are decomposed, and the correlation between the two markets is studied. It is shown that high frequency detail components represent much more energy than low-frequency smooth components. The relation between copper future daily returns in LME and that in SHFE is different on different time series scales. The fluctuations of the copper future daily returns in LME have large effect on that in SHFE in 32 days scale, but have small effect on high frequency scales. It is also evidenced that strong effect is existed between LME and SHFE for monthly responses of the copper futures but not for daily responses.(3) Propose a framework of anti-money laundering based on intelligent technology. Through the summarization of money laundering and anti-money laundering, the paper proposes a framework of anti-money laundering based on intelligent technology, including KYC customer due diligence, feature extraction of customer profiling, unusual behavior detecting and tracking technology. With these, users can build potential money laundering scenarios by rules-based systems, neural networkand data mining, et al, then analyse transactions to detect risk. In this paper, we describe the key technology in the four parts.(4) Use decision tree and nerve net methods to classify the risk status of customer. Money laundering (ML) involves moving illicit funds, which may be linked to drug trafficking or organized crime, through a series of transactions or accounts to disguise origin or ownership. China is facing severe challenge on money laundering with an estimated 200 billion RMB laundered annually. We apply decision tree method and BP method to create the determination rules of the money laundering risk by customer profiles of a commercial bank in China. A sample of twenty-eight customers with four attributes is used to induce and validate a decision tree method. The result indicates the effectiveness of decision tree in generating AML rules from companies' customer profiles.(5) Explore the use of statistical method and social network analysis in money laundering. With the users' transactions in Internet Banking, we can monitor the key customers who have suspicious money laundering dealing. Through the social network analysis and graph visualization, we obtain that many transactions including the fund transfer between the shops and persons. This information can help us monitor the suspicious effectively.