Addressing Vagueness and Uncertainty in the Banking Industry: Combining Fuzzy Modulatory Data Envelopment Analysis and Fuzzy Multivariate Regression

ABSTRACT This study develops a novel two‐stage framework to evaluate financial health, distress and liquidity traps in the banking industry by integrating a Fuzzy Modulatory Data Envelopment Analysis model with a Fuzzy Multivariate Lognormal Regression. The methodology incorporates fuzzy set theory with alpha‐cuts to capture vagueness and uncertainty in inputs, outputs and intermediate variables, addressing inherent data imprecision in financial systems. An empirical analysis of 93 Chinese banks (2014–2023) reveals key drivers of inefficiency, including labour costs, capital adequacy and market concentration, while highlighting the moderating effects of competition and regulatory measures. Results demonstrate improved efficiency trends and convergence in operational practices, influenced by regulatory compliance and macroeconomic policies. The framework further identifies contextual factors—such as the Herfindahl–Hirschman Index and CAMELS ratings—impacting inefficiencies under varying levels of uncertainty, offering a comprehensive risk management and policy evaluation tool. This research contributes to operational research by bridging gaps in banking efficiency analysis and providing actionable insights for enhancing financial stability and resilience under uncertain market conditions.