Real Estate Portfolio Investment Based On Semi-variance Risk Measurement Models

Currently,the real estate investment is showing a trend of diversification,with the national macro-control and real estate business risk continuing to increase.Investors could achieve the purpose of spreading risk by diversified investment between different projects in property types,geographical varieties and during different time energetically.Portfolio theory is the modern core theory to solve such problems and it's an important theoretical basis for investment decisions activities.The financial investment theory transitioned from descriptive qualitative analysis to the stage of scientific and rigorous quantitative analysis,because of the modern portfolio theory.The theory has been supported by numerous scholars since generated and has generated a lot of research.However,most study on portfolio risk measurement revealed irrationality and complexity in the actual operation gradually,which are always unsatisfactory in guiding the investment.For example,variance risk measurement actually consider the stability of income,the use of VAR does not have integrity.Based on the understanding of the original definition of the risk,this paper try to measure risk by semi-variance.The paper mainly research at the manner in which to measure the portfolio risk in the real estate decision-making,and how to determine the proportion of the portfolio for the investors with different risk tolerance.The paper consists of three parts.In the first part,it studies the real estate portfolio theory,the theory of risk and investor psychology,laid the foundation for the further study.In the second part,it analyze and compares several risk measures adequately,and selects the semi-variance to measure risk.It also proves the consistent between semi-variance risk measurement method and second-order stochastic dominance.In the third part,it introduces risk tolerance coefficient to improve the model by analyzing investors' risk tolerance,and solves the quadratic programming model by using Matlab procedures,then makes the empirical research.Empirical evidence shows that the semi-variance risk measure has more advantages,and the portfolio model based on the risk tolerance coefficient of the investors could reflect investor psychology,which has a better expression than the previous models.The model portfolio is carried out several discussions by introducing a multi-parameter,which improves the flexibility of the mode.Thus,it could guide the different investors to choose the right investment portfolio more effectively.