Financial Data Abnormality Alerting and Empirical Research on Chinese Real Estate Enterprises - based on Information Entropy and Logistic model

This paper summarizes the abnormal warning signal indicators of listed real estate enterprises' financial data and their behavioral characteristics. The study is based on 708 valid financial data of real estate enterprises from 2010 to 2022 in the Cathay Pacific database. The data was classified into normal (73) and abnormal real estate enterprises (121). The study screened and constructed the warning indicator system using information entropy and modeled it through Logistic regression analysis. The model passed the statistical test. The study shows that: (1)the warning model constructed by fusing information entropy and logistic regression achieves a warning accuracy of 88.70%. (2)Financial indicators such as accounts receivable turnover ratio, inventory turnover ratio, net profit margin, and return on assets play a significant role in identifying abnormal financial data effectively. Real estate enterprises should prioritize monitoring these indicators. (3)Compared with traditional financial analysis models, logistic regression analysis demonstrates higher accuracy and practical value in identifying financial anomalies of real estate enterprises. The study aims to offer a valuable reference for real estate companies and auditors to promptly identify and uncover potential financial anomaly risks.