基于复杂网络理论的股票指标关联性实证分析

摘要/Abstract

摘要： 复杂网络理论是研究股票市场内在结构和功能的有力工具,股票指标的关联性已成为研究股票市场的一个新视角。基于复杂网络理论对深沪300指数(2011年7月28至2012年2月28)构建网络拓扑结构。利用DFA进行数据筛选,计算筛选后数据的股票指标的绝对相关系数。根据股票指标的绝对相关系数概率分布确定阈值,构建网络拓扑结构,计算股票指标的网络统计特征指标,分析收益率、成交量、市盈率的网络结构。研究结果表明,股票收益率和成交量指标的具有较强的关联性,具有小世界性质;市盈率指标具有较弱的关联性,不具有小世界性质,是随机网络且受外界影响较大、效率低下,风险比较大。

关键词: 复杂网络, DFA, 股票指标, 拓扑结构

Abstract: The complex network theory is a powerful tool to study the internal structure and function of the stock market. The research of the relevance of stock indicators has become a new perspective in the research of stock market. Traditional financial research using equity-linked network to dig intrinsic clustering structure, lack of formulation of the relationship between the complex structures. This paper based on complex network theory to build the network topology on CSI 300 Index (July 28 2011-February 28 2012), the relationship between the complex structures is analyzed, the existing empirical analysis methods of complex network are improved. The data is tiltered by DFA, and the absolute correlation coefficient of the stock index is calculated. According to the absolute correlation coefficients of stock index probability distribution, the threshold is determined and build the network topology is build. Network statistics characteristic index of the stock is calculated, and the network structure of the rate of return, trading volume and price-earnings ratio are analyzed. The results show that the stock returns and trading volume indicators have strong correlation with the small-world character,while price-earnings ratio index has a weak correlation, and does not have the small-world character, and is a random network that subject to outside influence with low efficiency and high-risk. Such new tools and empirical results are useful to study the relevance and empirical findings of stock indicators in China's financial market.

Key words: complex network, the DFA, stock indicators, topology

中图分类号:

F830.9

张来军, 杨治辉, 路飞飞. 基于复杂网络理论的股票指标关联性实证分析[J]. 中国管理科学, 2014, 22(12): 85-92.

ZHANG Lai-jun, YANG Zhi-hui, LU Fei-fei. Empirical Analysis of Relevance of Stock Indicators Based on Complex Network Theory[J]. Chinese Journal of Management Science, 2014, 22(12): 85-92.

参考文献

[1] 徐红利, 周晶, 徐薇. 基于累积前景理论的随机网络用户均衡模型[J]. 管理科学学报, 2011, 14(07): 1-7.

[2] Caldareli G, Battiston S, Garlaschelli D, et al. Emergence of complexity in financial networks[J]. Lecture Notes in Physics, 2005, 650(9): 399-423.

[3] Albert R, Barabasi A L. Statistical mechanics of net works [J]. Reviews of Modern Physics, 2002, 74 (1): 47- 97.

[4] 徐有俊, 王小霞, 贾金金. 中国股市与国际股市联动性分析[J]. 经济经纬,2010, (05): 124-128.

[5] Kim H J. Kim I M. Scale-free network in stock markets[J]. Journal of the Korean Physical Society, 2002, 40(6): 1105-1108.

[6] 李亚静, 何跃, 朱宏泉.中国股市收益率与波动性长记忆性的实证研究[J]. 系统工程理论与实践, 2003, 23(01): 9-14.

[7] 汪小帆, 李翔, 陈关荣. 复杂网络理论及其应用[M]. 北京: 清华大学出版社, 2006.

[8] 何诚颖. 中国股市市盈率分布特征及国际比较研究[J]. 经济研究, 2003, (9): 74-81, 95.

[9] 颜竹梅, 刘亦文, 黄静寅. 中国股市市盈率的变化趋势及国际比较[J], 经济与管理, 2008, 22(5): 25-28.

[10] 赵静梅, 吴风云. 数字崇拜下的金融资产价格异象[J]. 经济研究, 2009, (6): 129-141.

[11] 陈浪南, 孙坚强. 股票市场资产收益的跳跃行为研究[J]. 经济研究, 2010, (4): 54-66.

[12] Kim H J, Lee Y, Kahng B, et al. Weighted scale free network in financial correlations [J]. Journal of Physical Society of Japan, 2002, 71 (9): 2133-2136.

[13] Onnela J P, Kaski K, KerteszJ. Clustering and information in correlation based financial networks [J]. The European Physical Journal B, 2004, 38 (2): 353-362.

[14] Lee K E, Lee J W, Hong B H. Complex Networks in a Stock Market [J]. Computer Physics Communications, 2007, 177 (1-2): 186.

[15] 黄玮强, 庄新田, 姚爽. 中国股票关联网络拓扑性质与聚类结构分析[J]. 管理科学, 2008, 21 (3): 95-96.

[16] Tibely G,Onnela J P,Saramki J,et al.Spectrum, intensity and coherence in weighted networks of a financial market. Physical A, 2006, 370(1): 145-150.

[17] Eom C,DHG,Kim S. Deterministic factors of stock networks based uncross-correlation in financial market. Physical A, 2007, (383): 139-146.

[18] Tabak B M, Serra T R, Cajueiro Do. Topological properties of stock market networks: The case of Brazil[J]. Physical A:Statistical Mechanics and its Applications, 2010, 389(16): 3240-3249.

[19] Tumminello M, Aste T, Matteo T D, et al. A tool for filtering information in complex systems [J]. Proceedings of the National Academy of Sciences of the United States of America, 2005, 102(30): 10421- 10426.

[20] Tumminello M, Matteo T D, Aste T, et al. Correlation based networks of equity returns sampled at different time horizons[J]. The European Physical Journal B, 2007, 55 (2): 209-217.

[21] 黄小原, 庄新田, 张泉. 股市波动的标度无关性算法及应用研究[J]. 管理科学学报, 2001,4(6): 55-59.

[22] 蔡世民,洪磊,傅忠谦,等. 基于复杂网络的金融市场网络结构实证研究[J]. 复杂系统与复杂性科学, 2011, 8 (03):29-33.

[23] Xian Yubo,Mei Lin. Adaptive expectation,complex network and the dynamic of standard diffusion—research based on computational economics [J]. Journal of Management Sciences, 2007, 20(4): 62-71.

[24] Barabási A L, Bonabeau E. Scale-free networks [J]. Scientific American, 2003, 288(5): 60-69.

[25] Fama E F, French K R. Common risk factors in the returns on stocks and bonds [J]. Journal of Financial Economics, 1993, 33(1): 3-56.

[26] Fuller R J, Huberts L C, Levinson M J. Returns to EIP strategies, higgledy-piggledy growth, analysts' forecast errors, and omitled risk factors [J]. The Journal of Portfolio Management, 1993, 19(2): 13-24.