Epidemic risk and corporate financing decisions: machine learning based prediction and ESG factor analysis

The recent worldwide health challenges, especially epidemics, have greatly influenced corporate funding choices, revealing weaknesses in conventional financial models. In this situation, companies are progressively incorporating Environmental, Social, and Governance (ESG) factors into their approaches to manage risk and bolster resilience. To handle risk and bolster resilience in this climate, businesses are integrating ESG considerations more and more into their strategies. To predict financial behavior in times of stress, this research employs machine learning (ML) techniques to explore the link between pandemic risks and financial decisions made by firms. This study introduces an ESG-Aware Investment Portfolio (ESG-AIP) model that utilizes Efficient Chicken Swarm tuned Extreme Gradient Boosting (ECS-XGBoost) for predicting stock returns and making corporate financing decisions in stressful circumstances. The research collects financial data and ESG metrics from companies during pandemic and post-pandemic eras, including stock values, debt amounts, and ESG ratings. Standard portfolio selection models assist in allocating assets to maximize returns while minimizing risk, duly considering constraints like ESG criteria. ECS aids in optimizing ESG elements, while XGBoost can be used for making more precise and robust stock return forecasts. The suggested ECS-XGBoost method has achieved an R2 of 0.8625, an RMSE of 1.03, a MAPE of 1.99, and a MAE of 0.54. This study highlights the critical role that ESG factors play in mitigating risk during crises and provides insights into how companies adjust their financing strategies in response to public health emergencies with an emphasis on long-term sustainability.