基于VAR—GM(1.1)—SVR模型的房地产价格预测研究

为提高房地产价格预测精度,克服传统统计数据真实性低、时效性差的缺点,本文以网络搜索数据为基础,首先通过斯皮尔曼相关分析和时差相关分析筛选出与房地产价格具有高度相关性的先行关键词,并利用向量自回归模型(VAR)和GM(1.1)模型分别预测房地产价格;然后构建基于向量自回归模型和GM(1.1)模型的VAR—GM(1.1)—SVR模型将以上两个模型的预测结果进行预测融合,并以西安市数据为例进行验证,得出均方误差(MSE)和标准平均方差(NMSE)分别为0.97和0.03,优于单一模型预测效果.

Research on Real Estate Price Forecast Based on VAR-GM(1.1)-SVR Model

In order to improve the accuracy of real estate price forecasting and overcome the shortcomings of traditional statistical data with low authenticity and poor timeliness,this paper based on network search data,firstly selects the first correlation key words with real estate price through Spearman correlation analysis and time difference correlation analysis,and using vector auto-regressive model(VAR)and GM(1.1)model to predict real estate prices respectively;then construct VAR-GM(1.1)-SVR model based on vector auto-regressive model and GM(1.1)model The prediction results of the model are predicted and merged,and the Xi’an data is taken as an example to verify that the mean square error(MSE)and standard mean variance(NMSE)are 0.97 and 0.03,which is better than the single model prediction effect.