Abstract
Purpose – This study investigates the application of machine learning (ML) techniques in predicting firm performance, responding to the challenges posed by the large volumes of data required for accurate predictions. It aims to assess the effectiveness of various ML methods and algorithms used in recent research, focusing on the prediction of firm performance across multiple dimensions.
Design/methodology/approach – A systematic literature review was conducted, examining 70 studies published over the last decade (2013–2023) that utilize ML techniques for firm performance prediction. This methodology allowed for an in-depth analysis of the attributes, methods, and algorithms commonly applied in the field, offering insights into the evolution and effectiveness of these approaches over time.
Findings – The research highlights the importance of considering a broad range of attributes beyond traditional financial metrics, such as financial health, market positioning, operational efficiency, innovation capability, leadership quality, and employee engagement, in predicting firm performance. It reveals a predominance of classification methods in ML, with neural networks, logistic regression, and decision trees being the most frequently employed algorithms. These findings underscore the potential of ML techniques to provide a more nuanced and accurate prediction of firm performance by integrating diverse data sources and attributes.
Practical implications – The study's insights have significant implications for investors, financial analysts, corporate management, policymakers, and regulators. By adopting a more comprehensive ML-based approach to performance prediction, these stakeholders can make more informed decisions regarding resource allocation, capital budgeting, investment strategies, and policy formulation. Improved predictability also aids in the development of more effective regulations and policies, benefiting the broader economic landscape.
Originality/value – This research contributes to the existing literature by systematically reviewing and synthesizing the application of ML techniques in firm performance prediction over a substantial period. It offers a consolidated view of the methods and attributes that are most effective in this context, highlighting the shift towards more complex and holistic approaches to understanding firm dynamics. This comprehensive overview provides valuable insights for future research and practice in the field of business analytics and performance prediction.