Forecasting corporate financial performance using sentiment in annual reports for stakeholders’ decision-making

Abstract

This paper is aimed at examining the role of annual reports’ sentiment in forecasting financial performance. The sentiment (tone, opinion) is assessed using several categorization schemes in order to explore various aspects of language used in the annual reports of U.S. companies. Further, we employ machine learning methods and neural networks to predict financial performance expressed in terms of the Z-score bankruptcy model. Eleven categories of sentiment (ranging from negative and positive to active and common) are used as the inputs of the prediction models. Support vector machines provide the highest forecasting accuracy. This evidence suggests that there exist non-linear relationships between the sentiment and financial performance. The results indicate that the sentiment information is an important forecasting determinant of financial performance and, thus, can be used to support decision-making process of corporate stakeholders.

Keywords:

• financial performance
• financial distress
• bankruptcy forecasting
• annual reports
• sentiment analysis
• opinion mining

JEL Classification:

• C45
• G33

Additional information

Notes on contributors

Petr Hajek

Petr HAJEK (assoc. prof. MSc. Ph.D.) is an associate professor with the Institute of System Engineering and Informatics, Faculty of Economics and Administration, University of Pardubice, Czech Republic. He has been working with the modelling of economic and financial processes using soft computing methods. He has published his research in leading journals such as Knowledge-Based Systems, Decision Support Systems, etc.

Vladimir Olej

Vladimir OLEJ (prof. MSc. Ph.D.) is a professor with the Institute of System Engineering and Informatics, Faculty of Economics and Administration, University of Pardubice, Czech Republic. He has been working with the modelling of economic and environmental processes on the basis of soft computing methods. He has published a number of papers concerning fuzzy logic, neural networks and genetic algorithms.

Renata Myskova

Renata MYSKOVA (assoc. prof. MSc. Ph.D.) is an associate professor with the Institute of Economics and Business Management, Faculty of Economics and Administration, University of Pardubice, Czech Republic. She has been working with the strategic management, management analysis, financial reporting and financial management. She has published a number of papers concerning economics and finance.