Abstract
Purpose
This study aims to explore the application of machine learning (ML) models in predicting financial distress within Taiwanese firms and examines their contribution to sustainable finance and climate resilience. By evaluating different ML models, this study aims to determine which models are most effective for early financial distress detection.
Design/methodology/approach
Seven ML algorithms were tested on a dataset of Taiwanese enterprises from 1999 to 2009. Models were assessed for predictive accuracy, with a focus on identifying those that excel in financial distress forecasting.
Findings
Random Forest and Gradient Boosting emerged as the top-performing models, outperforming traditional statistical methods in predictive accuracy. The study’s findings highlight the potential of these advanced models in the early detection of financially distressed firms, aiding sustainable investment.
Practical implications
By identifying high-risk firms early, these ML models promote proactive risk management and support capital allocation towards climate-resilient, environmentally responsible companies. This approach encourages the development of policies that align financial strategies with climate change objectives.
Originality/value
This research emphasizes the practical benefits of advanced ML approaches in financial distress forecasting, supporting sustainable finance and enhancing corporate alignment with climate change and sustainability goals.