The Intersection of Artificial Intelligence and Financial Modeling: Evaluating WACC Through Machine Learning and Generative AIDevaprasad, Maanya Artificial Intelligence (AI)Machine Learning (ML)Generative AI (GAI)Weighted Average Cost of Capital (WACC)Financial ModelingXGBoost AlgorithmChatGPTCost of EquityCost of DebtFinanceB S This thesis examines the application of artificial intelligence (AI) in financial modeling, specifically in calculating the weighted average cost of capital (WACC). This critical metric is used in the financial industry to analyze investment opportunities and company valuations. This study explores the efficacy of machine learning (ML) algorithms and generative AI (GAI) models in predicting WACC when providing key financial metrics and market factors. A comprehensive dataset from numerous academic databases such as CRSP, Compustat, and Fama-French was used to compute WACC through the traditional financial methodologies, establishing the study’s baseline, referred to as the raw WACC. The study's machine learning component employs the XGBoost algorithm, which the study selected based on its efficiency, ability to handle large datasets, and capability to detect patterns and predict WACC values. The model’s performance is gauged using mean squared error and R-squared values, which indicate substantial predictive accuracy and generalization. Similarly, the generative AI component utilizes ChatGPT to calculate WACC based on publicly available market data. However, the results from ChatGPT display variability in terms of the average deviation and a weak correlation metric. The comparative analysis highlights the ML model's accuracy and reliability, demonstrating the ongoing potential of algorithmic learning in relation to financial calculations. Furthermore, the study underscores the limitations of the current generative AI models in handling similar complex financial calculations. In general, this research contributes to the growing discourse on AI’s role in financial modeling, suggesting that ML algorithms can enhance the accuracy of financial predictions. On the other hand, generative AI models require further advancement to perform similar financial calculations. The study's findings emphasize the importance of integrating AI-driven methodologies to improve the efficiency and precision of financial metrics—such as WACC—in the broader financial industry.Stefan M Lewellen, Thesis SupervisorBrian Spangler Davis, Thesis Honors Advisoropen_access2025-03-20T01:29:06Zhttps://honors.libraries.psu.edu/catalog/9476mbd5817