基于FCM聚类的土地项目融资风险研究

Study of Risk Assessment on Land Program Financing Based on FCM Clustering

针对我国土地融资风险较大和土地储备部门风险评估的方法不够全面的问题,文章采用某省级土地储备中心从收到的土地储备项目可研报告中提取的数据,建立方差分析、模糊均值聚类的数学模型,得出构建项目投资总额估算、申请贷款额度、预期收益和申请贷款额度为评估体系,以风险系数三作为划分分位数,可以将储备项目按照储备风险划分为3个等级,并据此为土地储备融资提供参照依据。

In China, the risk of land financing is increasing, meanwhile land reserve department does not assess the risk in comprehensive way. According to the data exacted from land reserve project feasibility study report released by a provincial land reserve center, we established mathematical model of analysis of variance, fuzzy c -means clustering, to construct a evaluation system which included total project investment estimation, amount of loans, expected profit. By taking risk coefficient 3 as a division of quantile, projects were divided to 3 levels according to reserve risk, and thus to provide reference for land reserve financing.