EVALUATING MEASURES OF MARKET RISK IN CIRCUMSTANCES OF GLOBAL FINANCIAL CRISIS – EMPIRICAL EVIDENCE FROM FIVE COUNTRIES

Ivica Terzić () and Marko Milojević ()
Additional contact information
Ivica Terzić: Singidunum University
Marko Milojević: Singidunum University

CBU International Conference Proceedings, 2013, vol. 1, issue 0, 75-81

Abstract: The purpose of this paper is to evaluate performance of value-at-risk (VaR) produced by two risk models: historical simulation and Risk Metrics. We perform three backtest: unconditional coverage, independence and conditional coverage. We present results on both VaR 1% and VaR 5% on a one-day horizon for the following indices: S&P 500, DAX, SAX, PX and Belex 15. Our results show that Historical simulation 500 days rolling window approach satisfies unconditional coverage for all tested indices, while Risk Metrics has many rejection cases. On the other hand Risk Metrics model satisfies independence backtest for three indices, while Historical simulation has rejected more times. Based on our strong criteria to accept accuracy of VaR models only if both unconditional coverage and independence properties are satisfied, results indicate that during the crisis period all tested VaR models underestimate the true level of market risk exposure.

Keywords: value at risk; backtesting; market risk; historical simulation; risk metrics (search for similar items in EconPapers)
JEL-codes: C53 G17 G32 (search for similar items in EconPapers)
Date: 2013
References: Add references at CitEc
Citations:

Downloads: (external link)
https://ojs.journals.cz/index.php/CBUIC/article/view/17/19 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:aad:iseicj:v:1:y:2013:i:0:p:75-81

DOI: 10.12955/cbup.v1.17

Access Statistics for this article

More articles in CBU International Conference Proceedings from ISE Research Institute
Bibliographic data for series maintained by Petr Hájek ().