Abstract
This thesis explores ways to enhance stress-testing frameworks in commercial banks by integrating macroeconomic shocks and machine learning (ML) models in emerging markets. Based on an analysis of Basel III supervisory standards, Presidential Decree No. DP-5992, and scholarly literature, it is shown that while traditional linear stress tests are established, they fail to capture non-linear macro-financial tail risks and high-dimensional borrower default dynamics. A conditional sequence for implementation, the "macroeconomic shock vector – hybrid ML credit-loss engine – explainable AI (XAI) capital trigger" principle, and five readiness assessment indicators are substantiated.
References
1. Hull J.C. Risk Management and Financial Institutions. Hoboken: John Wiley & Sons, 2023. 832 p.
2. Financial Markets and Financial Technologies: textbook / S. Elmirzayev, N. Shavkatov, N. Sherkuziyeva, A. Karimov. Tashkent: Diadema Nur Servis, 2023. 412 p.
3. Jiyanova N. Financial technologies and their impact on financial markets // Economics and Innovative Technologies. 2022. №3. P. 9–17.
4. On the Strategy for Reforming the Banking System of the Republic of Uzbekistan for 2020–2025: Decree of the President of the Republic of Uzbekistan dated May 12, 2020 No. DP-5992. Tashkent: Adolat, 2020. 38 p.
5. Basel Committee on Banking Supervision. Stress testing principles. URL: https://www.bis.org/publications/201810-guidelines-stress-testing-principles (accessed: 16.10.2026).