基于多重分形的贵金属市场风险研究

The risk of precious metal market based on multifractal analysis

近些年来,由于金融危机、欧债危机的出现和通货膨胀的加重,具有避险保值功能的贵金属成为了更多投资者的投资对象.然而贵金属市场的价格波动是一个具有非线性特征的变化过程,分形理论为这种复杂的市场行为提供了一种新型的定量分析方法.本文对多重分形去趋势波动分析法(MF-DFA)和重叠平滑窗MF-DFA方法进行改进,提出了二分重叠平滑窗MF-DFA方法(Bi-OSW-MF-DFA)并且证明了此方法具有更强的稳健性和更高的效率,再运用改进的模型结合标度指数和多重分形谱对黄金和白银市场的多重分形特征和风险性进行研究.证实了贵金属市场的多重分形结构由长程相关性和厚尾概率分布两个因素共同决定,其中序列的波动性是造成贵金属市场具有多重分形特性的主要因素;并且发现了白银市场的多重分形强度较大,其风险高于黄金市场.

In recent years, with the aggravation of the global economic crisis and inflation, precious metals with safe-haven function have become increasingly popular. However, the price fluctuation of precious metals is a non-linear process, and multifractal theory has provided a new quantitative analysis method for these complicated market behaviors. In the present work, market Multifraetal Detrended Fluctuation Analysis （MFDFA method） and Overlapped Sliding Window-based MF-DFA （OSW-MF-DFA） are improved, generating a stronger model-Bi-OSW-MF -DFA method. Besides, the multifractal characteristics and risks of gold and silver market are analyzed in combination of the advanced model with scaling exponent and multifractal spectrum. It is found that a broad probability density function and different long-range correlations of small and large fluctuations are the two contributors for multifractality in precious metals market. Additionally, the risk of silver market is higher than the gold market.