Prediction of real estate price variation based on economic parameters

It is well known that many economic parameters may more or less influence the real estate price variation. In addition, the banker and investor are also interesting to know the real estate price future change. There had not appropriate model for including these factors for price prediction. Here, the influences of most macroeconomic parameters on real estate price variation are investigated before establishing the price fluctuation prediction model. Here, back propagation neural network (BPN) and radial basis function neural network (RBF) two schemes are employed to establish the nonlinear model for real estates price variation prediction of Taipei, Taiwan based on leading and simultaneous economic indices. Those prediction results are compared with the public Cathay House Price Index or the Sinyi Home Price Index. The mean absolute error and root mean square error two indices of the price variation are selected as the performance index. The public related data of Taipei, Taiwan real estate variation during 2005 ~ 2015 are adopted for analysis and prediction comparison.