Market risk: Exponential weighting in the value-at-risk calculation

Arbeitspapier

When measuring market risk, credit institutions and Alternative Investment Fund Managers may deviate from equally weighting historical data in their Value-at-Risk calculation and instead use an exponential time series weighting. The use of exponential weighting in the Value-at-Risk calculation is very popular because it takes into account changes in market volatility (immediately) and can therefore quickly adapt to VaR. In less volatile market phases, this leads to a reduction in VaR and thus to lower own funds requirements for credit institutions. However, in the exponential weighting a high volatility in the past is quickly forgotten and the VaR can be underestimated when using exponential weighting and the VaR may be underestimated. To prevent this, credit institutions or Alternative Investment Fund Managers are not completely free to choose a weighting (decay) factor. This article describes the legal requirements and deals with the calculation of the permissible weighting factor. As an example we use the exchange rate between Euro and Polish zloty to estimate the Value-at-Risk. We show the calculation of the weighting factor with two different approaches. This article also discusses exceptions to the general legal requirements

Object Details

• Language: Englisch

References and Relationships

• Bibliographic citation: Series: CEPIE Working Paper ; No. 04/20

Classification and Topics

• Classification: Wirtschaft ; Single Equation Models; Single Variables: Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models; Diffusion Processes ; General Financial Markets: Government Policy and Regulation ; Financial Institutions and Services: Government Policy and Regulation
• Subject: risk management ; market risk ; exponentially weighted moving average ; weighting scheme ; Value-at-Risk

Contributors, Places and Time

• Event: Geistige Schöpfung
• (who): Broll, Udo ; Förster, Andreas
• Event: Veröffentlichung
• (who): Technische Universität Dresden, Center of Public and International Economics (CEPIE)
• (where): Dresden
• (when): 2020

Further information

• Handle: http://hdl.handle.net/10419/224735
• URN: urn:nbn:de:bsz:14-qucosa2-720096
• Last update: 10.03.2025, 11:44 AM CET