Statistical Research On The Quality Of Accounting Information In The Annual Reports Of Listed Companies In My Country

Since listed companies have been required to disclose financial reports,accounting fraud,financial fraud and other phenomena have become increasingly fierce and repeatedly prohibited.With the continuous strengthening of regulatory measures by various regulatory authorities and regulatory agencies,the proportion of public financial fraud cases and the phenomenon of explicit and easily discovered “artificial manipulation” accounting information have decreased,but the phenomenon of implicit and difficult perceived “artificial manipulation” of accounting information has increased.The quality of accounting information is not only of great significance to enterprises,but also has a great impact on the development of social economy.Therefore,this paper uses statistical methods to study the accounting information quality of annual report data of Listed Companies in China.Firstly,through the calculation of the frequency of the first digits of the ten financial indicators of all listed companies in China in 2016 and 2017,it is found that the selected ten indicators are basically in accordance with the distribution of Benford's law.Through fitting testing,the indicators with better fitting effect are “income tax expense”,“business cost”,“operating income”,“total current assets” and “asset total”.“Management expenses”,“Net profit” and “total liabilities” are considered that there may be accounting information quality problems,among which “management expenses” are the indicators most likely to have quality problems.Then,according to the fitting testing results,select the main indicators as variables and take the real estate listed companies in the A-share main board market in China for example,build panel data models for relevant data for nearly ten years,and analysis the outliers of the model residuals,thus detecting a total of 45 data abnormal data and 9 abnormal sample companies.Finally,followed by a comparative analysis of major violations of listed companies,found that 7 out of 9 abnormal companies are indeed serious violations of the law,the correct rate reached 77.78%.In this paper,the Benford fitting test is carried out on the indicators before the modeling,then combined with the panel data model for empirical analysis.Which narrows the scope of financial indicators that may have the problems of accounting information quality and provides a new idea for the identification of audit work and accounting fraud to a certain extent.However,there are still some deficiencies need to improve in this paper.