基于ARMA模型的二维码安全保险资金流控

Two-Dimensional Code Security Insurance Fund Flow Control Based on ARMA Model

摘要: 针对二维码盗取个人信息及诈骗的问题，构建一款二维码扫码保险的模型。本文引入纳斯达克综合指数(NASDAQ)，采用自回归移动平均(ARMA)模型对未来资金流动情况实施监测，对市场的现状进行分析及未来发展预测，其中资金流动及使用的策略是此服务能否顺利实行的关键。通过分析不稳定数据进行差分处理得到的自相关和偏自相关图和对平稳性的检验，设定模型形式和滞后阶数进行建模。通过残差检验，得到的残差图是随机正态分布的、不自相关的，可见残差是一段白噪声信号，其有用的信号已经都被提取到ARMA模型中。结果表明，使用ARMA模型能够很好地对未来资金流动情况进行预测。

Abstract: To solve the problem of two-dimensional code theft of personal information and fraud, a two-dimensional code scanning insurance model is constructed. In this paper, NASDAQ composite index is introduced, and the ARMA model is used to monitor the future capital flow, analyze the current situation of the market and predict the future development. Among them, the capital flow and the strategy of use are the key to the smooth implementation of this service. The autocorrela-tion and partial autocorrelation graphs obtained by differential processing of unstable data are analyzed and the stationary is tested. The model form and lag order are set for modeling. According to the residual test, the residual graph obtained is randomly normal distribution and non-autocorrelation, indicating that the residual is a segment of white noise signal, whose useful signals have been extracted into the ARMA model. The results show that the ARMA model can pre-dict the future capital flow well.