Leveraging Machine Learning For Sustainable Investment: Evaluating The Predictive Power Of Esg Metrics On Funding Allocations

The integration of Environmental, Social, and Governance (ESG) factors into financial modeling hasbecome very crucial, as investors explore investment strategies for sustainable and responsible investment. Thisstudy probes how the use of Machine Learning (ML) techniques can be used to predict and allocate funding,based on ESG performance of an organization. The key objective is to evaluate the predictive accuracy ofdifferent ML models, in quantitatively forecasting funding allocations using ESG metrics. Dataset such as ESGmetrics and financial indicators including revenue and profit were collected from different companies. Toimprove the model’s predictability, feature engineering which involves creating new features that combine ESGmetrics with financial indicators to capture complex relationships affecting funding allocations was used. Theanalysis results indicate that the Random Forest model achieved the highest predictive accuracy, with an R² valueof 0.8657, demonstrating its robustness in managing complex, non-linear relationships within the dataset.Insights from the Feature Importance Analysis revealed that social and governance-related metrics, such asSocial Score Profit and Governance Score Profit, were the most significant predictors of funding received. Boththe Random Forest and Gradient Boosting ensemble methods outperformed traditional regression and SVMmodels in predicting funding allocations based on ESG performance, demonstrating their effectiveness inrevealing operational risks and opportunities. The findings from the study highlights the value ML models, suchas the ensemble methods, give in guiding investors towards companies with good social responsibility and ethicalpractices.