Research On Portfolio Strategies Of Systemic Risk Erasing Based On Price Volatility

In China’s stock market, stock volatility often occurs in a given period, which wouldbring individual investors unexpected losses. Markowitz portfolio theory laid the foundationof modern portfolio theory; however, portfolio can not do anything to avoid the systematicrisk but only reduce non-systematic risk. Taking Small and Medium Enterprise (SME) marketas the research object, this dissertation analyzed the emergence of minimum and maximumvalues of stock price volatility systematically. By researching Markowitz portfolio theory andstock price volatility theory, a series of regression models on phase-cycle minimum andmaximum values of stock price volatility were designed in prerequisite constraints, whichtook different phase stock costs as reference values. Point estimation and interval estimationwere used to test and compare the effectiveness and authenticity of predictive minimum andmaximum values of different models to verify the optimal models. Finally, the optimal modelswere applied to simulate stock investment operations to promote the countermeasures of stockselection and portfolio management. The proposed countermeasures based on the applicationof the optimal models can guide individual investors to deal with stock investment effectively.The main contents of this dissertation are as follows:1. Considered the usefulness of Markowitz portfolio theory and its reduction in avoidingthe systematic risk, a series of regression models on phase-cycle minimum and maximumvalues of stock price volatility were designed.2. Taking SME market as the research object, six regression models on phase-cycleminimum and maximum values of stock price volatility were designed. A variety of statisticaltests were used to determine the effective model initially. Then multiple regression modelswere revised for the multicollinearity problem.3. Using historical data, different prediction methods were applied to verify and analyzethe validity and accuracy of different models to draw the conclusion that the revised multipleregression models were optimal.4. Based on regression models on phase-cycle minimum and maximum values of stockprice volatility, three different modes of investment operations were proposed to simulatestock investment including the investment following promoted countermeasures, theinvestment without following the countermeasures and indexing. The empirical results showthat the investment portfolio using regression models to select stock and operate can get amuch higher rate of return than others, and can avoid systemic risk effectively.