基于银行流水数据的洗钱风险综合评估

Comprehensive Evaluation of Money Laundering Risk Based on Bank Flow Data

摘要: 针对洗钱账户的行为特点，在银行个人客户流水数据的基础上建立洗钱风险综合评价指标体系，利用因子分析方法和熵值法建立了洗钱风险综合评价模型，并利用广义极值分布对洗钱综合风险值的分布进行了拟合。采用重庆市某商业银行2018年个人流水数据进行了实证分析，其结果表明，该银行的绝大部分个人客户洗钱风险值都比较低，只有少数的客户可能有洗钱行为。根据拟合出的综合风险分布，在0.05的显著水平之下，有5239 (2.66%)位客户可能有洗钱风险，需要进一步的鉴别，这无疑将极大地提高银行的工作效率。

Abstract: According to the behavior characteristics of money-laundering, the comprehensive evaluation index system of money-laundering risk is established on the basis of the flow data of bank personal customers. The method of factor analysis and entropy method are used to establish the comprehensive evaluation model of money-laundering risk, and the generalized extreme value (GEV) distribution is used to fit the distribution of comprehensive risk value of money-laundering. The empirical analysis is carried out by using the customers personal flow data of a commercial bank in Chongqing in 2018. The results show that the risk value of money-laundering of most bank individual customers is relatively low, and only a few customers may have money-laundering behavior. According to the fitted comprehensive risk distribution, at the significant level of 0.05, there are 5239 (2.66%) customers who may have money-laundering risk, and they need further identification, which will undoubtedly greatly improve the efficiency of the bank.