基于神经网络和Z-score模型的公司财务预警
Financial Pre-Warning of Company Based on Neural Networks and Z-Score Model

作者: 马雯琦 :四川大学商学院,四川 成都;

关键词: 财务预警人工神经网络上市公司Financial Pre-Warning Artificial Neural Network Listed Companies

摘要:
文章基于2014、2015年我国沪深两市上市公司的样本数据,比较了Z-score模型、BP神经网络模型对我国上市公司的财务预警效果。结果表明,后者的预测准确率明显远远高于前者。文章的实证结果给了我们以下提示:一是在国外采用破产标准作为财务危机或财务困境标志的环境下,Altman提出的企业多变量财务预警模型Z-score判定模型,并不是很适用于我国企业的实际情况,如果直接用这个模型,预警误判率较高;二是人工神经网络在财务预警这一领域的潜力很大,具有深入探讨的价值。

Abstract: Based on sample data in years 2014 and 2015 for companies listed on Shanghai and Shenzhen stock exchanges, this paper compares the financial pre-warning abilities of Z-score model and BP neural networks model. It turns out that the latter model obviously provides significantly higher rate of accuracy than the former. The empirical results give us some suggestions as follows: firstly, in the environment where bankrupt standard is adopted as the symbol of financial crisis or financial dilemma of oversea listed companies, the company multi-variable financial pre-warning model, Z-score model, put forward by Altman, is not much suitable for domestic listed companies, which, if used directly, would be proved to have a high misdiagnosis rate; secondly, the artificial neural network has great potential in the financial pre-warning field, worthy of being studied deeply.

文章引用: 马雯琦 (2017) 基于神经网络和Z-score模型的公司财务预警。 现代管理, 7, 93-99. doi: 10.12677/MM.2017.73014

参考文献

[1] 朱荣. 企业财务风险评价与控制研究[M]. 大连: 东北财经大学出版社, 2008.

[2] Beaver, W.H. (1966) Financial Ratios as Predictors of Failure Empirical Research. Journal of Accounting Research Supplement, 4, 71-111.
https://doi.org/10.2307/2490171

[3] Ohlson, J.A. (1980) Financial Ratios and the Probabilistic Prediction of Bankruptcy. Journal of Accounting Research, 18, 109-231.
https://doi.org/10.2307/2490395

[4] Aziz, D., Emanuel and Lawson, G. (1992) Bank Prediction: An Investigation of Cash Flow Based Models. Journal of Management Studies, (12), 62-67.

[5] 周娟, 王丽娟. 基于现金流量指标的财务危机预警模型分析[J]. 财会通讯, 2005(12): 107-109.

[6] 周志华. 机器学习[M]. 北京: 清华大学出版社, 2016.

[7] Hornik, K., Stinchcombe, M. and White, H. (1989) Multilayer Feedforward Networks Are Universal Approximators. Neural Networks, 2, 359-366.

分享
Top