摘要
在传统有效市场分析框架下,市场流动性波动特征不容易进行准确刻画和分析,流动性预测有效性和准确性也无法得到充分保障。在分形市场假说下,利用多重分形去趋势波动分析法(MF-DFA)与多重分形去趋势交叉相关性分析法(MF-DCCA),研究我国股指期货与现货市场流动性、以及两个市场间相关性的非线性波动特征及其形成原因;随后,利用趋势熵维数(TED)、交叉验证优化支持向量机模型(CV-SVM)对股指期现货市场流动性波动情况进行智能预测;最后是稳健性检验。研究发现,我国股指期货市场流动性水平明显低于现货市场;股指期货市场与现货市场流动性及其相关性均呈现显著多重分形特征,这种多重分形特征均来源于长程相关性和厚尾分布。此外,股指期货市场流动性的多重分形程度显著高于现货市场,股指期货流动性的趋势一致性更显著,现货流动性的趋势逆反性更强;现货市场流动性的预测精度更高。
By performing price discovery and risk management functions,stock index futures provide a valuable hedging platform for the stock market.Good market liquidity not only guarantees that investors can complete transactions at expected prices in the stock market at low cost and quickly,but also provides a foundation for the proper functioning of stock index futures.As stock market liquidity directly affects its own effectiveness,the liquidity of stock index futures indirectly affects the stock market.There is an objective correlation between the stock and stock index futures market,for the stock index is the underlying asset of stock index futures,the volatility of the stock market directly affects the stock index futures market,and meanwhile the stock index futures play price discovery and risk management functions,which in turn indirectly affect the stock market.It is important to study the correlation between stock index futures and spot.Finally,the actual financial market does not conform to the traditional efficient market hypothesis and is actually a nonlinear market.This means that the behavior and dynamics of financial markets cannot be described simply by linear models.To study the liquidity of the stock index futures and spot markets,it is important to consider the specific characteristics of the relevant markets,especially the nonlinear characteristics,because these characteristics can significantly affect liquidity dynamics and price volatility.Meanwhile,a theoretical elucidation of the predictability of future fluctuations in liquidity is crucial.Only by theoretically illustrating the predictability of future fluctuations in liquidity can we provide an effective guarantee for the study of volatility trend prediction.The current research has paid little attention to the liquidity itself and the correlation between stock index futures and spot markets,and little research has been done on the fractal characteristics of liquidity and the prediction of future volatility trends,so this study can make up for this research gap,provide a new perspective for understanding the complexity of liquidity in stock index futures and spot markets.Through an analysis of the multifractal characteristics of liquidity in depth,market dynamics can be better grasped,which is crucial for investors as they can formulate more rational investment strategies based on the future trend of liquidity to obtain better investment returns.At the same time,for regulators,understanding the volatility of liquidity trends can help them better target the regulation of stocks that may have liquidity anomalies,thus maintaining the stability and healthy development of financial markets.In order to deeply analyze the volatility characteristics of liquidity in China’s stock index futures and spot market,reveal the nonlinear relationship and predict the future volatility of liquidity,under the framework of nonlinear fractal market,this paper firstly uses the historical trading data of CSI300 stock index futures and CSI 300 index,and with Amihud illiquidity index,chooses MF-DFA and MF-DCCA to study the characteristics and causes of the individual market liquidity and the liquidity correlation of stock index futures and spot market.Subsequently,the trend entropy dimensionality method and CVSVM are used to study the changing trends and predict future volatility of stock index futures and spot market liquidity.Finally,the above findings are tested for robustness.The results show that first,both the individual market liquidity and liquidity correlation of stock index futures and spot market exhibit significant multifractal characteristics and long memory,and multifractal characteristics are caused by the combination of long-range correlation and thick-tailed distribution;secondly,the degree of multifractality of stock index futures is significantly stronger than that of spot,and there is a significant inverse relationship between the maturity of stock index futures and the degree of multifractality;finally,stock index futures with stronger multifractality have higher strength in trend consistency but lower forecasting accuracy,while spot with lower multifractality has higher strength in trend inversion and higher forecasting accuracy.
作者
燕汝贞
张菁洋
吴栩
朱文龙
YAN Ruzhen;ZHANG Jingyang;WU Xu;ZHU Wenlong(School of Business,Chengdu University of Technology,Chengdu 610059,China;School of Management Science,Chengdu University of Technology,Chengdu 610059,China;School of Business,Qingdao University of Technology,Qingdao 266525,China)
出处
《运筹与管理》
CSSCI
CSCD
北大核心
2024年第12期217-223,I0097-I0100,共11页
Operations Research and Management Science
基金
中国博士后科学基金面上项目(2022M720545)
国家自然科学基金青年基金项目(71903017)
四川省自然科学基金面上项目(2023NSFSC0523)
成都市软科学项目(2023-RK00-00127-ZF)
成都市哲学社会科学规划项目(NY0520231285)
成都理工大学“双一流”建设哲学社会科学重点建设项目(ZDJS202201)。
关键词
流动性
多重分形
股指期货
股指现货
liquidity
multifractal
stock index futures
stock index spot
作者简介
燕汝贞(1982-),男,山东淄博人,博士,教授,博士生导师,研究方向:证券市场微观结构;张菁洋(1997-),女,四川绵阳人,硕士研究生,研究方向:证券市场微观结构;通讯作者:吴栩(1986-),男,四川巴中人,博士,教授,研究方向:分形理论;朱文龙(1985-),男,山东淄博人,博士,教授,研究方向:公司治理。
