Research on credit risk measurement and prevention in energy industry based on modified KMV-logit model

Under the backdrop of accelerating the global transition toward clean and low-carbon energy systems, the energy industry is confronting multifaceted credit risk challenges compounded by accelerated technological iteration, the phase-out of policy subsidies, and intensified industrial chain competition. To enhance credit risk early warning for energy companies, this study proposes an optimized hybrid evaluation model. Addressing the limitations of traditional KMV models in capturing asset return distribution and defining default points, the model integrates multi-parameter optimization techniques, a weighted averaging method, an EGARCH volatility process, and a genetic algorithm. It combines the dynamic discrimination capability of distance-to-default with the nonlinear classification strengths of the Logit model. Empirical tests on 106 Chinese energy listed companies from 2019 to 2023 show a 94% prediction accuracy. The study contributes to the literature by proposing a multi-parameter optimization framework that addresses the limitations of traditional models in capturing asymmetric volatility and static default points, thereby enhancing the precision of credit risk measurement in the energy sector.