Research On Portfolio Optimization Of Financial Markets Under Complex Interdependence

With the rapid development of the world economy and the rapid development of financial liberalization,the interdependence between financial markets has become complicated,and the diversification of financial investment and the continuous deepening of non-stop trading mechanisms have made the links between financial markets increasingly close.As a hotspot and a difficult point in the financial field,the cross-regional and cross-market investment portfolio has been closely watched by the theoretical and practical circles.Especially in the new situation of frequent financial crises,the close relationship between financial assets makes the portfolio The optimization is facing more severe challenges.Therefore,how to build a portfolio optimization model more effectively to reduce the risk of financial portfolio,thereby maintaining financial and economic security and promoting social harmony and stability,is not only an important task for the financial and economic management departments in the new situation,It is also a hot issue of academic concern.Since the 30 years of reform and opening up,China's economy has achieved world-renowned achievements.Various financial markets have gradually formed.From scratch,from small to large,the formation of multi-level transactions,diversification of trading venues and diversification of trading mechanisms have been formed.And other characteristics of the financial market system.Financial products in different markets are openly traded in different markets,and different trading mechanisms exist in different markets.Moreover,due to the difference in trading products and trading extremes,the complexity of China's financial market is further deepened,how to effectively clarify diversified financial The interdependence between markets undoubtedly puts higher demands on investors' investment decisions and policy formulation of financial regulators.Therefore,how to integrate multiple markets,analyze the interdependence between different financial markets in the financial market system from a macro perspective,and then explore the important theoretical and practical significance of portfolio risk and portfolio optimization.Based on this,in order to build a portfolio covering diversified financial markets,this article will be used for the SSE Bond Index,Shanghai and Shenzhen 300 Stock Index,Shanghai Stock Exchange Index,Shanghai and Shenzhen 300 Stock Index from January 1,2012 to December 1,2018.Futures,Shanghai Interbank Demolition Interest Rate,RMB Foreign Exchange Index,and Gold Spot Index are the research samples.Firstly,the sample data is processed by logarithmic rate of return,and ARMA-APARCH is used to characterize the "Stypical facts" of each financial market's rate of return series.At the same time,it combines the EVT theory to the edge distribution of the tail value of the yield series.Secondly,it is aimed at portfolio risk.In the depiction of the corresponding relationship,the vine copula selected by the Maximum Spanning Tree(MST)algorithm is used to characterize the interdependence structure between the seven financial markets,and then measure the VaR and CVaR under the vine copula dependent structure,and empirical comparison Back-testing was performed on the risk measures of five dependent structures,R-vine copula,C-vine copula,D-vine copula,R-vine all t,and R-vine all F.By comparing AIC,BIC,and maximum likelihood values,it was found that R-vine copula can more effectively and accurately fit the interdependence between diversified financial markets.The empirical analysis draws the following conclusions: First,the financial markets have complex interdependence structures,and the vine copula model can well describe the interdependence structure between financial markets,and the vine copula models C-vine and D-vine,which are relatively specific structures,Or R-vine all F,Rvine all t using a pair-copula function,R-vine copula has no specific structure,and the pair-copula function describing the nodes is also selected by the optimal criterion.Rvine copula is more effective at portraying the complex interdependencies between financial markets than several other vine copula models.Secondly,ARMA-APARCHSkewt can fit the financial income sequence well.By introducing the EVT extremum theory,the tail of the financial income sequence can be accurately characterized,and the tail information of the financial income sequence can be analyzed more effectively.The Mean-CVaR model based on the R-vine copula-dependent structure constructed in this paper can effectively optimize the portfolio.The optimization of the portfolio of the Mean-CVaR model based on the R-Vine Copula dependent structure is significantly better than the Mean-VaR model.The Mean-Variance model performs relatively poorly compared to the other two models.