test from 1990 to 2023

A comprehensive and sufficient database serves as a fundamental foundation for measuring and managing operational risk, the third major risk encountered by banks following credit risk and market risk. However, due to data privacy protection, the insufficient manual collection and delayed updates, etc., a publicly accessible operational risk database for risk analysis is currently lacking. This study proposes a new data collection framework by aggregating text mining methods and constructs one of the largest and most comprehensive external databases of cross-institution operational risk in China. Encompassing 3,723 operational risk events in Chinese commercial banks from 1990 to 2023, the database comprises 17 fields such as occurrence time, loss amount and business lines, providing a more comprehensive description of operational risk events. The loss severity and frequency of events within the database manifest characteristics that align with the findings in existing operational risk data research. We anticipate the constructed new database will contribute to advancements in operational risk capital calculations, event dependence analysis, textual analysis and institutional internal controls.