混沌递归分析在黄金市场风险识别中的应用

Financial crisis identification of Chinese golden market based on chaos recurrence theory

采用混沌递归图分析中国黄金市场的动态特征,实现黄金市场风险的识别和预测。采用混沌动力学分析法对中国黄金市场进行混沌性验证,实验结果表明中国黄金市场具有长期记忆及非线性特征,适合采用非线性动力系统模型进行建模研究。对黄金价格波动特征进行建模,定义了三类不同的特征区间。实现基于二维递归图的黄金波动特征时间序列的m维相空间重构及特征分析。采用递归压缩距离RPCD相似性匹配算法,实现黄金市场风险的识别。实验表明,提出的采用混沌递归实现黄金市场风险识别与预测,其误判率低于基于时域空间的欧式距离和动态时间规整的方法。

The method of chaos recurrence plot is used to analysis the dynamic properties of Chinese golden market and realizes crisis identification and prediction. The starting point for the study comes from chaotic dynamics analysis with the aim of testing whether the data from Chinese golden market may generate within the framework of a chaotic system.The results confirm that Chinese golden prices time series display long memory and can be modeled by the dynamical systems. In the second part, three different types of crisis phases among the golden market are come from. Then, the recurrence plot is developed to investigate the m-dimensional phase space trajectory through a two-dimensional representation of its recurrences. Finally, the similarity between recurrence plots is measured using RPCD distance to identify the crisis hidden behind the golden market. The empirical study shows that RP with RPCD similarity deduces the false positive effectively compared with Euclidean distance and Dynamic Time Warping.