Research On The Short-term Prediction Of Real Estate Price Based On Chaos Theory

With the constant change of the market economy, the real estate industry has become a pillar industry of the national economy. Real estate prices also became the focus of the society. The real estate price prediction is the basis and premise of the real estate industry healthy and orderly development. The real estate price prediction accurately and validly can not only provide reference for investment decision and the consumer decision-making, but also provide the basis to the relevant government department administration decision. Therefore, effective forecast for the real estate prices is particularly important.Exploring literature research, qualitative and quantitative combined, system dynamics, theoretical and empirical combined research methods, with the related theory of real estate prices and chaos theory as a foundation for analysis research. Firstly, introduces the real estate price prediction research at home and abroad, analysis the basic theory and chaos of the price real estate prediction, how to construct the chaos forecast and establish chaotic forecasting model. Then set up the chaotic forecasting model that based on the constructing principle of the real estate prices prediction model. The progress of real estate price system is non-linear and complicated in non-linear dynamics characteristics. The space reconstruction analysis of time series for real estate price added-value can make the entire real estate price system known more comprehensively. The chaotic characteristic of time series for real estate price is confirmed by reconstructing the space of time series, calculating the correlative dimensions and analyzing the maximum Lyapunov exponent. At last, to the residential housing price of Harbin for an example, and Application3-spline interpolation method to enlarge the sample number. Phase space reconstruction for the time series of real estate prices, taking the real estate price time series embedded in the reconstruction of phase space, fitting to the data and forecast by using the RBF neural network. The results show that the model is characterized by good accuracy and reliability of the prediction, which is a kind of effective methods to predict real estate prices.