Risk Contributions, Information and Reverse Stress Testing

Journal of Risk Model Validation, Vol. 3(2), (2009)

Posted: 28 Mar 2013

See all articles by Jimmy Skoglund

Jimmy Skoglund

SAS Institute Inc.

Wei Chen

SAS Institute Inc.

Date Written: March 25, 2009

Abstract

In this paper we describe methods of decomposing risk into subcomponents such as contributing instruments, subportfolios or underlying risk factors e.g., equity, foreign exchange, economy-wide systematic and interest rate risk factors. The Euler allocation principle for allocation of instrument and subportfolio risk contributions to the total portfolio risk is widely applied in practice for portfolio managers risk budgeting and economic capital allocations. However, the decomposition of risk factors into summable contributions is in general not possible and we describe a non-parametric method of extracting relative information of risk factors that is valid for general non-linear portfolios. The measure is based on the Kullback information theory and while it does not constitute a decomposition it can be used to judge the relative importance of risk factors in determining profit and loss. The measure is also useful for understanding the relevance of so-called reverse stress tests where a certain loss level is identified with certain risk factor values. The use of the information measure is illustrated with an application to a portfolio of derivatives.

Keywords: Risk contributions, capital allocation, risk information measures, reverse stress testing

Suggested Citation

Skoglund, Jimmy and Chen, Wei, Risk Contributions, Information and Reverse Stress Testing (March 25, 2009). Journal of Risk Model Validation, Vol. 3(2), (2009). Available at SSRN: https://ssrn.com/abstract=2239074

Jimmy Skoglund (Contact Author)

SAS Institute Inc. ( email )

100 SAS Campus Drive
Cary, NC 27513-2414
United States

Wei Chen

SAS Institute Inc. ( email )

100 SAS Campus Drive
Cary, NC 27513-2414
United States

Here is the Coronavirus
related research on SSRN

Paper statistics

Abstract Views
523
PlumX Metrics