中文摘要：

基于动态戈登模型构建包含住房市场收益率、租金增长率、利率、人均可支配收入增速在内的向量自回归模型,采用中国五个主要城市1999年第二季度至2011年第二季度的数据分析了住房市场收益波动的原因.结果表明：租金增长消息对住房市场收益波动的贡献程度最高,利率消息次之,两者之间存在的相关性减弱了市场收益波动;利率消息对住房市场收益的影响时间长于预期收益消息和租金增长消息;各城市住房市场收益对不同消息的响应强度和速度均存在差异性.

英文摘要：

The vector autoregression model consisting of housing market return, rental growth, interest rate and per capita disposable income growth was established based on dynamic Gordon model. And the reasons for return volatility in the housing market were analyzed with the data of five major Chinese cities from the second quarter of 1999 to the second quarter of 2011. The results indicate that the news of rental growth contributes the most to return volatility in Chinese housing market, followed by the news of interest rate, and their correlation weakens the volatility in the market return; the news of interest rate has longer impacts on return volatility than the news of expected return and rental growth; the responses of housing market return to various news in these cities show differences in terms of intensity and speed.