Amos Baranes, Rimona Palas
The prediction of earnings movement is used to evaluate corporate performance and make investment decisions. This study presents a detailed model for predicting the movement of company future earnings using Support Vector Machines (SVM) technique and comprehensive financial data extracted from the Securities Exchange Commission (SEC) mandated eXtensible Business Reporting Language (XBRL). XBRL does not change the disclosure itself; still it can facilitate information gathering and processing, since it is easily downloaded from the internet and translated into EXCEL format, which should be beneficial to users of the financial reporting information.
The model, using XBRL data, was able to classify the companies correctly on average, about 63.4% of the time, less than the traditional method of Stepwise Multivariate Logistic Regression which had an average prediction rate of 68.1%. However, when disseminating the data, based on industry, the accuracy level reaches 84.2% using SVM, while with Stepwise Multivariate Logistic Regression model accuracy only reaches 71.6%. These results suggest that the model presented has merit and can be used as a financial analysis tool.