Research On Robust Real Estate Portfolio Based On CVaR

With the rapid development of real estate industry and the continually rising of the housing price in China, there are many real estate investments in geographical and varieties portfolio among the real estate enterprises and investors. How to maximize the return on investment and minimize the risk of real estate investment portfolio is the main objective of the investors. In the face of a wide range of real estate investment options, as well as considering the lack of market data and the uncertainty of the parameters, it is strongly demanded by the rapid development of real estate industry that investors should determine what kind of products to choose and how to weigh the benefits and the risk of different products, so that to select the optimal investment portfolio, that is to avoid the underlying risk and achieve a stable benefit by the greatest possible degree.In the current macro economy and market environment, this paper gave full consideration to the complexity of the real estate market and the sensitivity of the model parameters, applied modern portfolio theory model to the real estate field, and proposed two robust real estate portfolio models, which gave a method of portfolio and risk control with quantitative analysis for the investors.This paper reviewed the modern portfolio theory and the return factor evaluation system, introduced CVaR risk measurement method and robust optimization method, and presented the real estate portfolio investment model based on CVaR. Then two robust models were established for the fact of the uncertainty of the return factors: considering the uncertainty of the parameters in the return vector based on the factor model, the first robust real estate portfolio model with risk measurement CVaR was developed and successfully transformed into a fixed model that can be solved effectively. This model can get the optimal solution and determine the optimal investment strategies when market parameters fluctuate randomly within the scope. Because of the policy dependency of the real estate market, the lack of data, and many other complex reasons, the probability of the return rate of portfolio could not be an exact one. So the second robust real estate portfolio model was set forward with the box and ellipsoid uncertainty sets of the return rate probability respectively. Also they are transformed into fixed models and effectively solved; the models would be more accurate and in line with market reality when applied into the real estate market. Furthermore,an extensive investment model including real estate and other financial properties portfolio was set up according to the former work.This robust real estate portfolio model provides more valuable theoretical support for the investment and will benefit the stabilization of the real estate industry. The last example shows that the real estate investment model provides better support and theoretical basis for portfolio selection.