摘要
为解决公租房REITs融资过程风险因素非线性、高度拟合特点等导致难以快速、高效对融资风险进行评价的问题,提出一种基于SVM-AFSA的公租房REITs融资风险评价模型。首先利用人工鱼群算法(AFSA)对支持向量机(SVM)的核函数及惩罚因子优化处理,寻找最优训练参数,提高支持向量机的预测精度;再借助支持向量机在处理非线性、小样本数据方面的优势对兰考县健康小区公租房REITs融资风险进行评价,结果表明该公租房REITs融资风险等级为一般。
In order to solve the problem that it is difficult to evaluate the financing risk quickly and efficiently because of the characteristics of public rental housing REITs financing risk factors with nonlinear and height fitting.This paper proposes a public rental housing REITs financing risk evaluation model based on SVM-AFSA.Firstly,AFSA is used to optimize penalty factor and the kernel function that supports vector machine,to find optimal training parameters and improve the prediction accuracy of support vector machine.Then,with the help of the advantages of the support vector machine in dealing with nonlinear and small sample data,the author evaluated the REITs financing risk about the Lankao County Health District public housing.The results show that the public rental housing REITs financing risk level is general.
出处
《建筑技术》
北大核心
2017年第7期768-771,共4页
Architecture Technology
基金
河南省哲学社会科学规划项目(2016BJJ051)
关键词
公租房
REITS
支持向量机
人工鱼群算法
风险评价
public rental housing
REITs
support vector machine
artificial fish swarm algorithm
risk evaluation
