|
|
|
| Analysts’ Identifying the Risk of Corporate Financial Fraud Based on Machine Learning |
| WU Bin，LIU Yunjing，ZHANG Min |
| 1. Renmin University of China， Beijing， China；2. Hunan University of Finance and Economics， Changsha， China |
|
|
|
|
Abstract Combining the risk of financial fraud prediction by machine learning method and the data of analyst recommendations, this study empirically investigates whether analysts identify the risk of corporate financial fraud. Using a sample of A-share listed companies from 2007 to 2018 for multiple regression analysis, we find that analysts tend to issue more negative rating reports for companies with higher risks of financial fraud, implying that analysts can identify the risk of corporates’ financial fraud and respond effectively during their information interpretation processes. This association is more prominent among analysts with more experience, higher reputation, or smaller conflicts of interest, suggesting that analysts’ ability and incentives drive analysts’ identifying the risk of corporate financial fraud. The analysis of the economic consequences of analysts issuing negative recommendations shows that analysts’ negative recommendations significantly reduce the probability of corporate financial fraud in the future.
|
|
Received: 28 June 2021
|
|
|
|
|
|
|
