Research On Multi-stage Portfolio Decision Based On Stochastic Programming And Decision Rules

Portfolio has been the darling of financial research field since it came out,and is favored by researchers and investors.In recent years,with the gradual improvement of China’s economic strength and the continuous improvement of people’s living standards,people’s mood to enter the investment market is constantly rising.Income and risk always show an associated relationship,and investment should be cautious when entering the market with risks.How to avoid or minimize risks is the concern of investors,and it is also a problem to be solved by portfolio.Portfolio mainly embodies the idea of diversification of investment,which can achieve the purpose of diversification of risks by choosing different risk assets,and at the same time,it needs a good asset allocation to make investors get corresponding benefits.Moreover,more and more investors tend to invest in multiple stages,which enables investors to make appropriate strategic adjustments in the face of unexpected situations.The risks in investment mainly come from various uncertainties,such as the rate of return.In multi-stage investment,how to express this uncertainty is particularly important.Existing literature mainly uses stochastic programming theory to deal with multi-stage portfolio problems,and uses scenario tree to express multi-stage uncertainty,which has achieved good results.In this paper,scenario tree is used to express the three-stage uncertain return rate,and decision rules are used to determine the nodes in the scenario tree.According to the stochastic programming theory,Markowitz’s single-stage mean-variance portfolio model is extended to multi-stage,and three multi-stage portfolio models with different objective functions are given.Through empirical analysis,the results show that the model is effective and can guide investors to invest.Combined with the results of empirical analysis,this paper analyzes the investment decisions of investors,and puts forward some suggestions for investors.The main contents of this paper include the following four aspects:Firstly,aiming at the uncertainty of the return rate,this paper forecasts the return rate based on ARMA-GARCH model,and uses Monte Carlo method to simulate the possible scenarios,thereby constructing a multi-stage scenario tree to express the uncertain scenarios.Secondly,after analyzing the mean-variance model proposed by Markowitz,this paper gives a decision rule to determine the nodes in the scenario tree,and establishes a multi-stage portfolio model with maximizing returns,minimizing risks and maximizing returns and minimizing risks by using stochastic programming theory.Thirdly,the model is empirically analyzed by using the index data of six stocks in China’s securities market,and the results show that the model is feasible and effective.It also visualizes the investment weights,investment returns and investment risks of investors with different temperaments in three stages and different periods,giving a simple and clear visual guidance to investment decision makers.Fourthly,some suggestions are given to investors.Investors should strengthen their ability to obtain relevant information,and be able to identify and analyze the information,and then use relevant technologies to predict the stock return rate.Moreover,conservative,steady and aggressive investors should consider their own situation when investing,and should not invest blindly.