Financial Stress Scenario Development in a Data-Rich Environment - a Practitioner's View

• Activity Number: 274 - Macroeconomic Forecasting and Policy in Data Rich Digital Age Environments
• Type: Invited
• Date/Time: Tuesday, July 30, 2019 : 8:30 AM to 10:20 AM
• Sponsor: Business and Economic Statistics Section
• Author(s): Xin Wang*
• Companies: HSBC/ IHS Markit
• Keywords: Stress testing; scenario design; market risk; Basel II; data science

Abstract

Since the 08 financial crisis, stress testing exercise has been recognized by both regulators and banks as an important means to assess the resilience of bank portfolios under adversarial scenarios. As big banks are exposed to hundreds of thousands, if not millions of risk factors, building a full scale stress scenario involves analysis and modelling over a huge amount of data. In this paper, we presents a practitioner's guide to stress scenario design in big data context. Borrowing recent progress in data science, such as the development of pandas and scikit-learn, we outline a Python based scenario design work- ow that accommodates scalable data storage, visualization, analytics and modelling. We apply the framework to the traded risk stress testing scenario requested by UK banking regulator Prudential Regulation Authority (PRA).