|本期目录/Table of Contents|

[1]马丰宁,宋明原. 基于离散化贝叶斯网络的财务预警研究 ——以香港上市公司为例[J].天津大学学报(社会科学版),2017,(3):198-203.
 Ma Fengning,Song Mingyuan. Financial Prediction Study Based on Discrete Bayesian Network —Take Listed Companies in Hong Kong for Instance[J].Journal of Tianjin University(Social Science),2017,(3):198-203.
点击复制

 基于离散化贝叶斯网络的财务预警研究

——以香港上市公司为例
()
分享到:

《天津大学学报(社会科学版)》[ISSN:1008-4339/CN:12-1290/c]

卷:
期数:
2017年3期
页码:
198-203
栏目:
现代企业管理
出版日期:
2017-05-15

文章信息/Info

Title:
 Financial Prediction Study Based on Discrete Bayesian Network

—Take Listed Companies in Hong Kong for Instance
作者:
 马丰宁 宋明原
 天津大学管理与经济学部
Author(s):
 Ma Fengning Song Mingyuan
 College of Management and Economics, Tianjin University, Tianjin 300072, China
关键词:
 贝叶斯网络 离散化 财务预警
Keywords:
 Bayesian network discretion financial prediction
文献标志码:
A
摘要:
 将贝叶斯网络应用到香港上市公司财务预警,结合基于线性相关性的启发式方法建立了6个不同相关性阈值对应的模型,将ESM离散方法应用于财务预警,并将其与已被多次使用的EPT离散方法从不同角度进行对比。结果表明:单模型角度下, ESM方法的表现优于EPT方法;而在模型联系角度下,当阈值不断降低,两种方法总体上均能提升模型预测的准确率。由于提升程度不同,综合考虑两种分析角度的结果后提出猜想并得到证实,与目的结点相关性系数绝对值在005以上的指标用ESM方法离散效果更好;与目的结点相关性系数绝对值在005以下的指标用EPT方法离散效果更好。
Abstract:
 Combined with heuristic method based on linear correlation, this paper applies Bayesian network to financial prediction among listed companies in Hong Kong and builds six different models corresponding to different correlation thresholds. We also apply the ESM discretion method to financial prediction and compare it with the EPT discretion method which has been adopted widely from various angles. The result shows that from the angle of single model, generally, the performance of the ESM method is better than the EPT method; but from the angle of linked models, with correlatio thresholds reduced gradually, generally speaking, the two methods can raise accuracy rate of prediction. Nevertheless, due to the different degree, we suggest a hypothesis after comprehensive deliberation from the abovementioned angles: the ESM method is the better choice for variables having the absolute correlation coefficients greater than 005 with purpose node; the EPT method is the better choice for variables having the absolute correlation coefficients smaller than 005 with purpose node. Finally, we adopt the mixed discretion method to verify the hypothesis.

相似文献/References:

[1]李健,张吉辉.水足迹视角下区域水资源灾变的灰色拓扑预测[J].天津大学学报(社会科学版),2013,(1):1.
[2]李秀芳,景珮. 基于多目标规划的最优再保险策略[J].天津大学学报(社会科学版),2013,(1):5.
[3]高素英,郝晓华,刘建朝,等. 滨海新区城市进化评价研究[J].天津大学学报(社会科学版),2013,(1):15.
[4]郭俊华. 京津冀文化消费水平实证研究[J].天津大学学报(社会科学版),2013,(1):21.
[5]赵黎明,卢 珊. 科技企业孵化器与创投合作绩效研究[J].天津大学学报(社会科学版),2013,(1):26.
 [J].Journal of Tianjin University(Social Science),2013,(3):26.
[6]张念木,胡连兴,郭建欣. 水电工程后评价的综合评价方法研究[J].天津大学学报(社会科学版),2013,(1):31.
 [J].Journal of Tianjin University(Social Science),2013,(3):31.
[7]姜姣娇,赵 涛. 电信业集团客户CRM价值评估模型设计与应用[J].天津大学学报(社会科学版),2013,(1):41.
 [J].Journal of Tianjin University(Social Science),2013,(3):41.
[8]李义丹,周志刚. 人的全面发展观与人的科学发展观关系研究[J].天津大学学报(社会科学版),2013,(1):46.
 [J].Journal of Tianjin University(Social Science),2013,(3):46.
[9]王 坤. 新农村建设中的非物质文化遗产之命运探析——以南乡义成永画店为例[J].天津大学学报(社会科学版),2013,(1):50.
 [J].Journal of Tianjin University(Social Science),2013,(3):50.
[10]段志成,陈 通,杨秋波. 灾后基础设施重建中公众参与的行为框架及路径分析[J].天津大学学报(社会科学版),2013,(1):54.
 [J].Journal of Tianjin University(Social Science),2013,(3):54.

更新日期/Last Update: