Research On The Optimal Asset Allocation Strategy Based On Stock Pairing

With the rapid development of the domestic financial market,the stock market has been great-ly shaken and affected by multiple factors,the financial stocks have not performed well.In this context,more and more people are concerned about reasonable and effective investment approach,in which asset allocation due attention.Asset Allocation of how wealth is allocated between the different asset classes,can be effective in dispersing risk,while also able to obtain significant benefits,which is a narrow portfolio theory.Based on this theory,this paper will take the optimal allocation of matching assets as the research object without considering the cost of margin financ-ing,and the asset allocation model of two kinds of risk measures(variance and quadratic variation)is designed for stock holders,and the corresponding Hamilton-Jacobi-Bellman equation(HJB)is obtained by using dynamic programming principle.In addition,in order to make pairs trading strategy more in line with the reality of the financial world,we also consider the stop check sur-plus and leverage constraints,which is one of the innovation of this article points.In the end,the numerical solution of the optimal strategy is obtained by using the wide stencil method,and then make a sensitivity or empirical analysis of it.The main research results are as follows.Firstly,we study the optimal investment strategy of cointegrated assets using the classical mean-variance portfolio selection criterion.That is,we desires to find controls which generate Pareto optimal points.And then the corresponding HJB equation model obtained by dynamic programming principle.In the last,accurate efficient frontiers are constructed by using wide sten-cil method and we show some numerical sensitivity analysis for the market parameters,namely the correlation p,the leverage constraints pmax.Secondly,we prove that the numerical scheme developed in the three chapter converges to the viscosity solution of the nonlinear HJB value equation.Finally,we using second variation as the risk measure,research the problem of optimal in-vestment strategy for risk assets,and the spread was considered as a O-U process.From an investor's point of view,with the greatest benefits for the target to establish the optimal control model,the corresponding HJB equation by dynamic programming principle.By using wide sten-cil method,then we achieve an optimal strategy.In numerical analysis we obtain main parameters in the model by least square estimate and maximum likelihood estimate,and filter out "New Chi-na Life Insurance" in Chinese A-share and H-share market.As a consequence,we demonstrate allocation strategy and invest profits by using 250 trading days' stock closed price data.