Study On The Method Of Real Estate Investment Risk Prediction Based On Support Vector Machine

Along with high-speed development of our national economy, there are more and more money and capital put into the real estate circle for its abundant profits earnings. However, risk always goes with any investment on the real estate, especially the laws and rules of our real estate market are incomplete at present. Once the risks occur, it will bring huge damages even bankruptcy to the investors. Therefore, it is an important topic for theoretical researchers and industrialists to develop an effective and accurate method to predict this inevitable risk and thus decrease the losses.Support vector machine (SVM) is a new machine-learning method based on statistics learning theory (SLT), and can deal with problems of classification and regression successfully. And the SVM method has become the hot pot in the academic fields for its excellent learning ability. Based on the imperfections of the traditional methods predicting the investment risk of real estate, we try to make full use of the excellent nonlinear fitting ability of SVM to make up for these imperfections. The main achievements are as follows:Firstly, the uncertain influential factors of real estate investment were analyzed and the risk analyses were finished; secondly, dentally expound the theory of SLT and SVM; Thirdly, the basic mechanism of using SVM to predict the risk of real estate investment was built up, its realization steps were established and some key problems which were very important in building SVM model were given; When case study, using a real example to demonstrate the realization of this model. Case study turned out that this model was feasible and of practical use.The dissertation provides a practicable method for the real estate investment risk prediction, which is important for our real estate circle to improve the level of investment decision-making.