基于稀疏主成分的我国上市公司信用风险评价与预测

把我国2016-2018年沪深Ａ股上市公司中164家ST公司作为信用违约样本，492家非ST上市公司作为非违约样本进行实证研究.从营运能力、偿债能力、盈利能力和成长能力等4个方面选取了25个财务指标，然后运用稀疏主成分方法提取主成分因子，并加入公司规模、第一大股东持股比例和股权质押3个非财务指标，作为Logistic回归模型的输入参数.在此基础上构建Logistic模型进行信用风险评价和预测.

Credit Risk Assessment and Prediction of Listed Companies in China Based on Sparse Principal Component Analysis

We conduct empirical research by taking 164 ST companies listed in Shanghai and Shenzhen A-shares from 2016 to 2018 as credit default samples and 492 non-ST listed companies as non-default samples. 25 financial indicators were selected from the four aspects of operating ability, debt paying ability, profitability and growth ability. Then, principal component factor was extracted by sparse principal component method, and three non-financial indicators including company size, shareholding ratio of the largest shareholder and equity pledge were added as input parameters of Logistic regression model. On this basis, Logistic model was constructed for credit risk assessment and prediction.