Method for Housing Price Forecasting based on TEI@I Methodology

Based on the TEI@I methodology proposed by Wang, et al, this paper presents an approach to forecast housing price. 114 indicators are selected by rough set theory, and the leading indicators are selected with time difference correlation analysis. Seasonal housing prices are forecasted by regression and grey models, and integrated via the wavelet neural network approach for error correction. Our analysis predicts that national commercial housing sales price would rise 6.88% in Q4-2006 and 6.64% in Q1-2007. Next, standard event study methodology is used to measure the effect on real estate investment of government policy, one of the most important indicators to forecast the housing price. It is found that the Chinese government's macro-policy in 2005 suppressed the growth of real estate investment and housing prices.