基于主成分回归分析的西安市房地产市场价格走势

Xi＇ an real estate market prices tendency based on principal component regression analysis

研究以西安市为例，选取影响房地产价格的人均GDP X1总人口X2、城镇居民人均可支配收入墨、土地购置费咒、房屋造价墨等11个可以量化的因素，采用SPSS软件进行主成分分析，将11个因素转化为3个主成分，然后建立模型，采用回归分析方法对3个主成分与各个年份房地产价格的关系进行分析，并对2012--2015年房地产均价进行预测。从预测结果来看，预测结果实际值与预测值的差异较小、精度较高，预测值可以在一定程度上反映西安市房地产市场状况，预测模型及预测结果能够为房地产企业以及政府、购房者的决策提供参考。

The resrarch takes Xi＇ an as an example, and selects GDP per capita X1 , total population X2 , the per capita disposable income of urban residents X3,1and acquisition costs X4, housing cost X5 and e- leven quantified factors which affect real estate prices. Models are built through principal components a- nalysis by using SPSS software. We transfers eleven factors into three main components, after which the eleven factors shall be built as a model by adopting regression analysis to analyze the ralationship be- tween three main components and the real estate prices each year, and then forcast the average price from the the year of 2012 to 2015. The actual value and predicated value has the small difference from the prediction results. To some extent, predicated value can reflect the real estate market condition of Xi＇ an. Predictive model and outcomes can provide reference for the real estate enterprise, goverment and decesion-making of the homebuyer.