Real Estate Prices Forecasting Based on Gray Correlation Analysis and Neural Network Model

In order to find out the main causes of reai estate price' changes, the gray correlation analysis was used to quantify the influence degree of each factor of real estate price. In order to modify the residuals of GM (1, 1) model, the gray neural network is established. First, the simulated values and residuals of real estate price sequence are produced by GM (1, 1) model, then, put the residuals which produced by model GM (1, 1) as inputs to the neural network. The effectiveness of our methodology was verified with an empirical study that compared GM (1, 1) model with the hybrid approach. And the results show that this method can be an effective tool to predict the real estate price, the precision of this method is advantage to GM (1, 1) model, which is useful to provide a scientific basis for the Marco control of government, investment decisions of real estate developers and purchase strategy of consumer.