摘要
为解决先验数据有限且存在大量不确定因素情况下,城市轨道交通周边房地产价格的预测问题,提出一种基于BP神经网络与马尔可夫链的组合预测模型。首先,采用BP神经网络,使用较少量的样本数据完成城市轨道交通周边房地产价格曲线的粗略拟合;在此基础上,借助马尔可夫链进行系统状态划分,缩小预测区间以提高预测精确度;最后,运用基于BP神经网络与马尔可夫链的组合模型,对北京市轨道交通13号线周边房地产价格进行了预测分析。计算结果表明,该模型具有较高的精度和可靠性。
An integrated model based on the neural network and Markov chain was established to predict the real estate price along the urban rail transit under the condition of limited a priori data and a host of uncertainties. A BP neural network was used to fit roughly the real estate price curve along the urban rail transit based on a little sample data. Markov chain was applied to achieve the state transition probability matrix of the system to modify the predicted result. The model takes the randomness of the real estate price into account, and it stll available regardless of that the statistics of the real estate prices are scarce. The real estate price along the rail transit No. 13 in Beijing was taken as an example to illustrate the application of the model. The calculation results showed that the model is characterized by good accuracy and reliability of the predicition.
出处
《吉林大学学报（工学版）》
EI
CAS
CSCD
北大核心
2008年第3期514-519,共6页
Journal of Jilin University:Engineering and Technology Edition
基金
“973”国家重点基础研究发展计划项目(2006CB705500)
中国人民大学科研基金项目(30307104)
关键词
交通运输系统工程
城市轨道交通
房地产价格预测
BP神经网络
马尔可夫链
engineering of communications and transportation system
urban rail transit
real estate
price
forecast
BP neural network
Markov chain
