从美国房价走势透视中国房地产市场走向

The Chinese Real Estate Market Outlook Based on the U.S.Housing Price Analysis

文章从量化分析的角度出发,研究了全美50个州和哥伦比亚特区房价指数序列的记忆性特征,发现各州房价指数序列均具有一定程度的长记忆性。同时,我们对50个州房价指数记忆参数的估计值进行了聚类分析,由此发现各州房价指数的记忆性呈现出一定的地区聚集性特征,而且还与地区的人口密度存在一定关系,从而为宏观政策制定者在决定房价调控措施的时候提供了有益的新思维,可使楼市更加健康稳定地发展。此外,文章的研究结果对于解决我国房地产市场目前存在的巨大泡沫问题提供了一个全新的视角。

In this paper we explore the long-memory features in the HPI time series of U.S.50 states from the perspective of quantitative analysis.The semi-parametric approach is utilized to estimate the memory parameter d of the HPI series and a cluster analysis is implemented according to the values of d.From the clustering results,we perceive a regional clustering phenomenon as well as a relationship between the values of d and the geographic locations（also,population density）,and hence it provides a useful reference to the policy makers when they launch the measures to rein in property prices,that is,the measures should vary from regions to regions in order to maintain a healthy and stable property market.The results of this study may be also conductive to reducing and preventing the real estate market bubbles or the abrupt plummet of house prices.