基于两类Boosting的财务造假识别方法对比
The Comparison of Financial Fraud Identification Methods Based on Two Types of Boosting
摘要: 针对财务造假问题，使用XGBoost和CatBoost两类boosting方法进行财务造假识别。对于超参数选择，通过GridSearchCV对两类boosting方法预训练，寻找最优超参数。将超参数应用于两类boosting方法上，从时间、准确度和性能的角度对比分析方法识别效果。结果表明，CatBoost方法相比于XGBoost方法，财务造假识别准确率更高，性能更优。
Abstract: In order to solve the financial fraud problem, XGboost and CatBoost methods are used to identify financial fraud. For the selection of super parameters, GridSearchCV is used to pre-train the two kinds of boosting methods to find the optimal super parameters. Super parameters are applied to two types of boosting methods, and the recognition effects of the methods are compared and analyzed from the perspectives of time, accuracy and performance. The results show that the CatBoost method has higher accuracy and better performance than the XGBoost method in identifying financial fraud.
文章引用：范宇晨. 基于两类Boosting的财务造假识别方法对比[J]. 应用数学进展, 2021, 10(12): 4227-4233. https://doi.org/10.12677/AAM.2021.1012449

参考文献

[1] Altman, E. (1968) Financial Ratios, Discriminant Analysis, and the Prediction of Corporate Bankruptcy. Journal of Finance, 23, 589. [Google Scholar] [CrossRef] 
[2] Messod, B.D. (1999) The Detection of Earnings Manipulation. Financial Analyst Journal, 5, 22-36. [Google Scholar] [CrossRef] 
[3] Tutino, M. and Merlo, M. (2019) Accounting Fraud: A Literature Review. Risk Governance and Control: Financial Markets and Institutions, 9, 8-25. [Google Scholar] [CrossRef] 
[4] Qin, R. (2021) Identification of Accounting Fraud Based on Support Vector Machine and Logistic Regression Model. Complexity, 2021, Article ID 5597060. [Google Scholar] [CrossRef] 
[5] Zheng, Y., Ye, X. and Wu, T. (2021) Using an Optimized Learning Vector Quantization-(LVQ-) Based Neural Network in Accounting Fraud Recognition. Computational Intelligence and Neuroscience, 2021, Article ID 4113237. [Google Scholar] [CrossRef] [PubMed]
[6] Tu, Z. (2005) Probabilistic Boosting-Tree: Learning Discriminative Models for Classification, Recognition, and Clustering. 10th IEEE International Conference on Computer Vision (ICCV’05), Vol. 1, 1589-1596.
[7] Chen, T. and Guestrin, C. (2016) Xgboost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining, New York, 13-17 August 2016, 785-794. [Google Scholar] [CrossRef] 
[8] Prokhorenkova, L., Gusev, G., Vorobev, A., et al. (2017) CatBoost: Unbiased Boosting with Categorical Features. arXiv preprint arXiv:1706.09516
[9] Paleczek, A., Grochala, D. and Rydosz, A. (2021) Artificial Breath Classification Using XGBoost Algorithm for Diabetes Detection. Sensors, 21, 4187. [Google Scholar] [CrossRef] [PubMed]
[10] Thongsuwan, S., Jaiyen, S., Padcharoen, A., et al. (2021) ConvXGB: A New Deep Learning Model for Classification Problems Based on CNN and XGBoost. Nuclear Engineering and Technology, 53, 522-531. [Google Scholar] [CrossRef] 
[11] Shahriar, S.A., Kayes, I., Hasan, K., et al. (2021) Potential of Arima-Ann, Arima-Svm, Dt and Catboost for Atmospheric Pm2. 5 Forecasting in Bangladesh. Atmosphere, 12, 100. [Google Scholar] [CrossRef] 
[12] Zhang, F,. Yang, J., Liang, B.S., et al. (2021,) Analysis of Influencing Factors of New Energy Vehicle Satisfaction Based on Scenario Thinking and Catboost Model. IOP Conference Series: Earth and Environmental Science, 769, 042023. [Google Scholar] [CrossRef] 
[13] Bao, Y., Ke, B., Li, B., et al. (2020) Detecting Accounting Fraud in Publicly Traded US Firms Using a Machine Learning Approach. Journal of Accounting Research, 58, 199-235. [Google Scholar] [CrossRef] 

为你推荐