How FinTech affects financial sustainability: Evidence from Chinese commercial banks using a three-stage network DEA-Malmquist model

Battle Count: 82

The paper is primarily focused on banking sustainability and FinTech impact assessment rather than trading strategies. However, findings about bank financial sustainability, loan efficiency, and profitability could inform credit risk models, sector rotation strategies, and macro-prudential trading signals. The DEA-Malmquist framework could potentially be adapted for efficiency-based stock selection in the banking sector.

The three-stage network DEA-Malmquist model requires solving multiple linear programming problems with complex inter-stage linkages. The additive efficiency decomposition method (Cook et al., 2010) adapted for three stages is non-trivial. The Malmquist index computation requires efficiency scores across multiple time periods and technology frontiers. The two-way fixed effects panel regression with IV/CF methods adds further complexity. Requires specialized DEA software or custom LP solvers.

The paper uses publicly available databases (CSMAR, Wind, China Statistical Yearbook, Peking University Digital Financial Inclusion Index, China National Intellectual Property Administration). The three-stage network DEA-Malmquist model is based on Shi et al. (2025) methodology. However, no code or specific implementation details for the DEA model are provided. The two-way fixed effects regression is standard econometrics. Data sources are identified but raw data access may require subscriptions.