Research on Identification of Fraud in Financial Reporting Based on Adaptive Group Lasso-Logistic Regression

摘要

Non-financial indicators are an important basis for fraud Identification in financial reports, but the quantification of non-financial indicators will also cause problems such as more dummy variables in the identification model, the correlation of explanatory variables and the decrease of identification rate, this makes it in the financial report fraud Identification application is more limited. In view of this situation, this paper proposes a fraud Identification model based on Lasso-Logistic regression of adaptive group, and selects the data of listed companies in China from 2010 to 2018 as a sample, considering both financial and non-financial indicators, an empirical study is carried out by using the model. The results show that this model has higher recognition rate and robustness than the previous Logistic model, Lasso-Logistic model and adaptive Lasso-Logistic model when several non-financial indicators are introduced, it solves the problem of the application of non-financial index in fraud Identification effectively, and has certain application value.