| 信用风险模型的贝叶斯改进研究 |
| |
| 引用本文： | 程建,连玉君,刘奋军.信用风险模型的贝叶斯改进研究[J].国际金融研究,2009(1). |
| |
| 作者姓名： | 程建 连玉君 刘奋军 |
| |
| 作者单位： | 北京大学光华管理学院;中山大学岭南学院;国家开发银行陕西省分行; |
| |
| 摘 要： | 基于小样本数据和外部先验信息,本文运用贝叶斯(Bayes)估计量来改进信用风险模型的违约预测力。同时,运用中国上市公司财务数据,分别对贝叶斯估计量和标准Logit估计量进行了模拟估计,并通过统计量AUC值和布莱尔分数(Brier Score)对其预测精度进行比较。结果表明,贝叶斯估计量具有更高的预测精度和稳定性。
|
| 关 键 词： | 信用风险模型 经验贝叶斯估计量 近似贝叶斯估计量 AUC值 布莱尔分数 |
A Study on Bayesian Improvement on Credit Risk Models |
| |
| Cheng Jian Lian Yujun Liu Fenjun.A Study on Bayesian Improvement on Credit Risk Models[J].Studies of International Finance,2009(1). |
| |
| Authors: | Cheng Jian Lian Yujun Liu Fenjun |
| |
| Abstract: | Based on the prior information of external credit risk models and the internal models built for predicting default by banks,this paper suggests two Bayesian estimators to improve the predictability of credit risk models. Therefore,we make a bootstrap simulation to compare the accuracy of Bayesian estimators and straight Logit estimators by AUC and Brier score. The result shows Bayesian estimators have a higher accuracy than straight Logit estimator. |
| |
| Keywords: | Credit Risk Model Empirical Bayesian Estimator Approximate Bayesian Estimator AUC Brier Score |
| 本文献已被 CNKI 等数据库收录！ |
