Multifractal spectrum analysis of nonlinear dynamical mechanisms in China’s agricultural futures markets
Multifractal spectrum analysis of nonlinear dynamical mechanisms in China’s agricultural futures markets
复制标题
DOI：
10.1016/j.physa.2009.12.009
复制
发表时间：
2010-04
影响因子：
3.3
通讯作者：
Shu-Peng Chen;Ling-Yun He
中科院分区：
文献类型：
--
作者：
Shu-Peng Chen;Ling-Yun He
基于配分函数和多重分形谱分析，我们研究了中国农产品期货市场的非线性动力机制，即大连商品交易所（DCE）和郑州商品交易所（ZCE），这两个市场几乎所有的农产品期货合约都在这两个市场进行交易。1（简称SB） DCE期货市场。其次，通过对原始时间序列进行洗牌，破坏了潜在的非线性时间相关性；因此，我们发现，在多重分形的市场形成中，长程相关机制是主要的贡献。第三，通过对左右半谱的演化跟踪，我们发现在价格波动较大的左半谱中存在临界点，在临界点之间存在不同的行为；但对于小价格波动的右侧谱，其宽度随着延迟t的长期增加而缓慢增加。最后，大波动的动态与小波动的动态明显不同，这意味着市场中多重分形的形成存在不同的潜在机制。
Based on Partition Function and Multifractal Spectrum Analysis, we investigated the nonlinear dynamical mechanisms in China’s agricultural futures markets, namely, Dalian Commodity Exchange (DCE for short) and Zhengzhou Commodity Exchange (ZCE for short), where nearly all agricultural futures contracts are traded in the two markets. Firstly, we found nontrivial multifractal spectra, which are the empirical evidence of the existence of multifractal features, in 4 representative futures markets in China, that is, Hard Winter wheat (HW for short) and Strong Gluten wheat (SG for short) futures markets from ZCE and Soy Meal (SM for short) futures and Soy Bean No.1 (SB for short) futures markets from DCE. Secondly, by shuffling the original time series, we destroyed the underlying nonlinear temporal correlation; thus, we identified that long-range correlation mechanism constitutes major contributions in the formation in the multifractals of the markets. Thirdly, by tracking the evolution of left- and right-half spectra, we found that there exist critical points, between which there are different behaviors, in the left-half spectra for large price fluctuations; but for the right-hand spectra for small price fluctuations, the width of those increases slowly as the delay t increases in the long run. Finally, the dynamics of large fluctuations is significantly different from that of the small ones, which implies that there exist different underlying mechanisms in the formation of multifractality in the markets. Our main contributions focus on that we not only provided empirical evidence of the existence of multifractal features in China agricultural commodity futures markets; but also we pioneered in investigating the sources of the multifractality in China’s agricultural futures markets in current literature; furthermore, we investigated the nonlinear dynamical mechanisms based on spectrum analysis, which offers us insights into the underlying dynamical mechanisms in China’s agricultural futures markets.