基于多维灰色系统理论的房价预测
|
摘要
房地产行业快速发展的同时也存在房价居高不下的现象,对于房价未来的发展趋势预测已成为学术研究和社会各界关注的重点。为了对房地产市场有进一步的了解,将灰色理论用于房价的预测问题,根据2007年—2016年全国房价及其作用成分的统计数据,运用灰色关联分析方法确定能够干扰全国房价的主要解释变量,包括:住房租金类居民消费价格指数、用于住宅类土地价格、房地产开发企业竣工房屋造价。结合GM(1,1)模型,对全国房价先模拟后预测,得出全国房价未来的发展走向,并比较二者的预测性能。
With the rapid development of the real estate industry,there is still a phenomenon of high housing prices. The prediction of the future development trend of housing prices has become the focus of academic research and social attention. To get a better understanding of the real estate market,the gray theory is applied to the prediction of house prices. Based on the statistical data of house prices and their components from2007 to 2016 in China,the main explanatory variables which can interfere with the national house prices are determined by the gray correlation analysis method,including the housing rent consumer price index and the use of in the residential land prices,real estate development enterprises completed housing costs. Combined with GM( 1,1) model,the national housing prices are simulated first and then predicted,and the future development trend of the national housing prices is obtained. The forecasting performance of the two models is compared to prove the accuracy of the model.
With the rapid development of the real estate industry,there is still a phenomenon of high housing prices. The prediction of the future development trend of housing prices has become the focus of academic research and social attention. To get a better understanding of the real estate market,the gray theory is applied to the prediction of house prices. Based on the statistical data of house prices and their components from2007 to 2016 in China,the main explanatory variables which can interfere with the national house prices are determined by the gray correlation analysis method,including the housing rent consumer price index and the use of in the residential land prices,real estate development enterprises completed housing costs. Combined with GM( 1,1) model,the national housing prices are simulated first and then predicted,and the future development trend of the national housing prices is obtained. The forecasting performance of the two models is compared to prove the accuracy of the model.
引文
[1]刘向东.我国房地产市场形势分析与展望[A].中国经济分析与展望(2017—2018)[C]. 2018:16.
[2]马海涛,陈琳,路正南.基于灰色理论的中国房地产价格指数预测[J].统计与决策,2007(19):117-118.
[3]李星瑶.房地产市场现状与政府调控政策分析[J].中国国际财经(中英文),2018(2):9.
[4]欧廷皓.基于ARMA模型的房地产价格指数预测[J].统计与决策,2007(14):92-93.
[5]邱启荣,于婷.基于主成分分析的BP神经网络对房价的预测研究[J].湖南文理学院学报(自然科学版),2011(3):24-26,36.
[6]高玉明,张仁津.基于遗传算法和BP神经网络的房价预测分析[J].计算机工程,2014,40(4):187-191.
[7]闫鹏飞,王典,燕慧慧.基于GM(1,1)模型的郑州市商品房房价预测[J].重庆交通大学学报(社会科学版),2013,13(3):60-63.
[8]王倩,王莹,邱继勤.基于灰色GM(1,1)和BP神经网络的房价预测[J].时代金融,2012(36):332-333.
[9]侯普光,乔泽群.基于小波分析和ARMA模型的房价预测研究[J].统计与决策,2014(15):20-23.
[10]尤梅芳,黄敏,程立. ARIMA模型在房价预测中的应用——四川省商品住房价格指数未来走势的实证分析[J].中国物价,2009(6):38-40.
[11]胡六星,吴结飞.基于灰色系统理论的两种房价预测方法比较[J].统计与决策,2010(7):22-24.
[2]马海涛,陈琳,路正南.基于灰色理论的中国房地产价格指数预测[J].统计与决策,2007(19):117-118.
[3]李星瑶.房地产市场现状与政府调控政策分析[J].中国国际财经(中英文),2018(2):9.
[4]欧廷皓.基于ARMA模型的房地产价格指数预测[J].统计与决策,2007(14):92-93.
[5]邱启荣,于婷.基于主成分分析的BP神经网络对房价的预测研究[J].湖南文理学院学报(自然科学版),2011(3):24-26,36.
[6]高玉明,张仁津.基于遗传算法和BP神经网络的房价预测分析[J].计算机工程,2014,40(4):187-191.
[7]闫鹏飞,王典,燕慧慧.基于GM(1,1)模型的郑州市商品房房价预测[J].重庆交通大学学报(社会科学版),2013,13(3):60-63.
[8]王倩,王莹,邱继勤.基于灰色GM(1,1)和BP神经网络的房价预测[J].时代金融,2012(36):332-333.
[9]侯普光,乔泽群.基于小波分析和ARMA模型的房价预测研究[J].统计与决策,2014(15):20-23.
[10]尤梅芳,黄敏,程立. ARIMA模型在房价预测中的应用——四川省商品住房价格指数未来走势的实证分析[J].中国物价,2009(6):38-40.
[11]胡六星,吴结飞.基于灰色系统理论的两种房价预测方法比较[J].统计与决策,2010(7):22-24.