نوع مقاله : مقاله پژوهشی
عنوان مقاله English
نویسندگان English
Abstract
The present study examines and compares the accuracy and efficiency of machine learning algorithms to predict the type of audit opinion in companies admitted to the Tehran Stock Exchange. In this regard, the method of variable selection (test comparing the average of two samples) has been used to examine and select the variables influencing the type of audit opinion. In order to achieve this goal, 1,606 company-years (146 companies for 11 years) of observation collected from the annual financial reports of companies admitted to the Tehran Stock Exchange during the period of 1390 to 1400 have been tested. In this research, out of six machine learning algorithms (decision tree and regression, random forest, neural network, nearest neighbor, logit regression, support vector machine) and also the findings of machine learning techniques show that the overall accuracy of decision tree techniques and Regression, random forest, neural network, nearest neighbor, logit regression, support vector machine are 76.3%, 77.7%, 76.9%, 74.6%, 78.3% and 69.6% respectively, which shows the efficiency of logit regression algorithm compared to other algorithms. In general, the error rate of the model in identifying companies with an acceptable audit opinion is 0.251, and for random forest, neural network, nearest neighbor, logit regression, and support vector algorithms,
کلیدواژهها English