Research On Financial Market Linkage And Financial Risk Contagion Under Complex System Theory

With the trend of financial liberalization and the development of economic globalization,the interaction and penetration of finance and economic activities,as well as the transmission and exchange of market information,the interaction and correlation between financial markets also show a significant upward trend.The interaction of financial markets has promoted the optimal allocation of global financial and economic resources,but it has also led to the widespread spread and deterioration of the frequent financial crisis in various financial entities in the world in recent years.The economic and financial development of all countries in the world is closely linked,whether it is the global financial system composed of the financial markets of each country,or a country or even a region,many financial individuals in the financial system have inextricably complex relationships,and ultimately constitute a complex financial system of different sizes.Therefore,with the spread of the financial crisis all over the world,the research on the correlation of financial markets and assets plays an important role in financial risk management,portfolio optimization and asset pricing.Considering that the financial market is a complex dynamic system,based on the complex system theory,this paper studies the financial market from three aspects: fractal analysis theory,complex network theory and random matrix theory.In this paper,the theoretical research and analysis are carried out,and the theoretical mechanism of complex system and financial market is studied.The concept and established characteristics of financial market interrelation are defined,as well as the application of complex network theory in the study of cross-correlation.After the introduction of the theoretical mechanism of the correlation between investor emotion and stock market return,then explains the relationship between complex network and financial risk in detail,and finally expounds the basic concepts and related methods of fractal analysis theory,random matrix theory and complex network theory.Then the cross-correlation measurement of financial market,the contagion of financial risk and portfolio optimization are empirically studied.Based on fractal analysis theory,complex network theory and random matrix theory,this paper studies the correlation between investor emotion and financial market,the correlation of multiple financial markets,the spread of international financial market risk,the correlation in financial market and the optimization of investment portfolio.First of all,based on fractal analysis,the correlation between the two financial markets is studied.Here we mainly carry out two aspects of research: on the one hand,the use of multifractal to study the correlation between investor sentiment and stock market returns in China is an effective combination of behavioral finance theory and financial market theory,and the theoretical application of fractal theory to the field of behavioral finance is a new attempt to fractal theory.In practice,it is helpful for government institutions to have a better understanding of investors,so as to optimize the structure of investors,improve the composition of the main body of market investment,and achieve the purpose of maintaining the stability of the financial market.On the other hand,the moving average cross-correlation of multifractal downward trend is adopted.The method of analysis is to study the cross-correlation between the capital market and the foreign exchange market,the commodity market and the gold market,respectively.It is a global study of the cross-correlation of financial market in China,and through the study of the cross-correlation between the capital market and the other three markets,the cross-market conduction mechanism of financial risk is grasped,and the conduction of financial risk is controlled globally.Therefore,using the fractal analysis theory to analyze the crosscorrelation of the financial market,it is found that the cross-correlation of the financial market presents the non-linear characteristic of the fractal and the multi-fractal,which further enriches the theory of the fractal market hypothesis.And the shortcomings of the traditional measurement and statistical methods in the cross-correlation measurement are also made up.Secondly,the cross-correlation in the financial market is studied based on the random matrix theory.In this paper,we combine the fractal analysis theory with the theory of DCCA and the theory of random matrix to study the cross-correlation between Chinese market stocks during the financial crisis and to analyze the statistical characteristics of cross-correlation under different time scales.At the same time,the random matrix theory is adopted to filter the random noise of the cross-correlation matrix of the financial assets,and the effect of the random matrix optimization method is compared,and the risk of the investment combination is optimized to some extent.This will help investors to build the portfolio model,choose the best combination strategy,and reduce the risk of the portfolio Insurance.Finally,the cross-correlation of multiple financial markets is studied based on the complex network theory.This chapter mainly studies the global stock market network from two aspects: one is to combine the whole international financial market as the research object,to combine the complex network and the method of the minimum spanning tree,tostudy the international financial risk conduction mechanism and the evolution of the network topology structure at different stages;The second is to study the cross-correlation of financial market from the angle of different time scales,and then to analyze the cross-market conduction of financial risks in the financial network under different time scales.Combining the Complex Network Theory with the Theory of Financial Risk Communication,the Study of the spread of the melting risk in the global stock market in the whole network perspective.By combining the qualitative analysis with the quantitative positive analysis,the interaction behavior and characteristics of the whole system are analyzed from a new perspective,so that the cross-correlation among the multiple markets can be effectively measured,which is not realized by the traditional cross-correlation measurement method.Studying the effect of the structure characteristics of the global stock market network on the financial risk transmission,shall be useful to understand the transmission mechanism and the characteristics of the financial crisis in the stock market,and how to develop the policy prevention and mitigate the spread of the financial risk,and as a result,good for maintaining the economic safety of the self.Due to the limitation of time and knowledge,there are still some problems in this paper,which is still not deep enough,is worthy of further improvement and supplement in the future research.For example,deeply study the following aspects: how to effectively combine the multifractal method with the classical financial theory to study and measure the nonlinear characteristics of the financial market,and the plane maximum filtering graph method and the threshold method in the complex network theory are used to study the correlation between multiple financial markets,so as to further mine the other information implied in the complex financial network,how to effectively estimate the theoretical distribution of eigenvalues of random matrices under different time scales,and quantitative study of RMT denoising method and so on.