摘要
以汪寿阳等(2005)提出的TEI@I方法论为指导,提出房价预测的研究框架.根据研究框架中对小样本数据的处理方法,首先基于粗糙集理论对114个影响房价的指标进行筛选,采用时差相关分析得到先行指标,分别建立回归模型和灰色模型预测季度房价,最后用小波神经网络进行误差校正,得到2006年4季度和2007年1季度全国商品房销售价格将分别同比增长6.88%和6.64%.由于房地产投资是预测房价的重要指标,文中以"国八条"为例用标准事件分析法分析政策对房地产投资的影响,得到"国八条"对房地产投资和房价上涨均有显著的抑制作用.
Based on TEl@I methodology proposed by Wang et al. （2005）, this paper proposes a housing price forecasting method. According to how to analyze small sample data in the research method, 114 important indicators are selected by rough sets theory, and the leading indicators are selected with time difference correlation analysis. Seasonal housing price is forecasted by regression model and grey model, integrated by wavelet neural network for error correction. The empirical study shows that the housing price will rise 6.88% in Q4-2006 and 6.64% in Q1-2007. Standard event study methodology is used to measure the political impact on the real estate investment, which is one of the most important indicators to forecast the housing price. The empirical analysis shows that macro-policy in 2005 suppressed the growth of the real estate investment and housing price.
出处
《系统工程理论与实践》
EI
CSCD
北大核心
2007年第7期1-9,51,共10页
Systems Engineering-Theory & Practice
基金
北京市科委博士生毕业论文资助基金(ZZ0518)
