【摘要】 选取2017年我国房地产上市公司的财务与非财务数据,构建行业信用风险评价指标体系,运用因子分析法计算各公司的综合得分Z值,利用K-Means进行聚类分析,确定信用评级基准。以2017年的样本数据为基础建立BP神经网络,将上市房企评级结果作为目标输出值,训练获得精度最优网络。套用该模型对2018年我国上市房企的信用评级进行仿真测试,拟合优度达0.734。研究结果表明:在近年经济下行与房地产市场宏观调控背景下,上市房企整体信用水平下降;上市房企的信用评级越优,其市场表现越好。因此,投资者可将上市房企的信用评级作为有效参考标准,以降低投资风险。更多还原

【Abstract】 In order to study the credit rating of listed real estate enterprises in China, the authors construct the industry credit risk evaluation index system with the financial and non-financial data of China ’ s listed real estate companies in 2017. The factor analysis is used to calculate the Z-score of each enterprise.The Z-scores of listed real estate enterprises in the sample set are classified by KMeans clustering analysis and they are regarded as China’ s listed real estate enterprises’ credit rating in 2017. Then, the BP neural network is established based on the sample data of this year, and the credit rating results of listed housing enterprises’ are taken as the target output value to train the BP neural network with the optimal precision. Using this model to conduct a simulation test on the credit rating of listed real estate enterprises in China in 2018, the goodness of fit reaches 0.734. The result shows that the overall credit level of real estate enterprises declines under the background of economic downturn and macroeconomic regulation of the real estate market. Besides, it also proves that those real estate enterprises with better credit ratings are more likely to have better market performances. Therefore, investors can take the real estate enterprise credit rating as an effective reference standard to reduce investment risk.更多还原