Leveraging artificial intelligence models for financial forecasting: A detailed analysis of predictive performance and benchmarks

Download This Article

This work is licensed under a Creative Commons Attribution 4.0 International License.

Abstact

The increasing integration of artificial intelligence (AI) into financial forecasting has garnered significant interest within finance and accounting domains. AI systems, proficient in mathematical and logical computations, can enhance the precision of financial predictions, making them invaluable for decision-making and risk management. This research investigates the predictive power of various AI models in forecasting critical financial metrics — specifically revenue and net income — while correlating their benchmark scores with predictive effectiveness. Previous literature underscores AI’s advantages over conventional statistical methods, with deep learning and ensemble approaches frequently cited for their accuracy in forecasting financial outcomes. Benchmark assessments such as the Multi-task Language Understanding (MMLU) and Grade School Math 8K (GSM8K) are integral in evaluating a model’s mathematical capabilities while problem-solving benchmarks like the ARC-Challenge and Graduate-level Problem Solving Questions (GPAQ) test their reasoning abilities. Despite these advancements, a comprehensive evaluation of AI models’ predictive accuracy across a variety of financial indicators still needs to be more extensive. This study addresses this research gap by analyzing historical data from ten publicly traded companies spanning 2020 to 2022 and predicting their 2023 financial performance. A zero-shot prompt-based approach is employed, and the predictive outputs are compared against actual financial results, assessing model accuracy in relation to benchmark scores. The findings of this study enhance the understanding of how AI can be leveraged for financial forecasting and provide practical insights for implementation in accounting practices. Emphasizing the importance of data quality, model transparency, and bias management, this research contributes to the growing body of knowledge on the application of AI in financial analysis.

Keywords: Artificial Intelligence, Financial Forecasting, Predictive Analytics, Benchmarking, Revenue Prediction, Net Income Prediction

JEL Classification: C45, C53, G17, M41

Received: 08.10.2024
Accepted: 23.10.2024

How to cite: Akpan, M. (2025). Leveraging artificial intelligence models for financial forecasting: A detailed analysis of predictive performance and benchmarks. In M. Pazarskis, A. Kostyuk, V. Santolamazza, & P. Capuano (Eds.), Corporate governance: Scholarly research and practice (pp. 96–100). Virtus Interpress. https://doi.org/10.22495/cgsrapp18