Value‐at‐risk under extreme values: the relative performance in MENA emerging stock markets

Abstract

PurposeThe paper aims to investigate the relative performance of the most popular value‐at‐risk (VaR) estimates with an emphasis on the extreme value theory (EVT) methodology for seven Middle East and North Africa (MENA) countries.Design/methodology/approachThe paper calculates tails distributions of return series by EVT. This allows computing VaR and comparing the results with Variance‐Covariance method, Historical simulation, and ARCH‐type process with normal distribution, Student‐t distribution and skewed Student‐t distribution. The paper assesses the performance of the models, which are used in VaR estimations, based on their empirical failure rates.FindingsThe empirical results demonstrate that the return distributions of the MENA markets are characterized by fat tails which implies that VaR measures relies on the normal distribution will underestimate VaR. The results suggest that the extreme value approach, by modeling the tails of the return distributions, are more relevant to measure VaR in most of the MENA.Research limitations/implicationsThe results show that the use of conventional methodologies such as the normal distribution model to estimate the financial market risk in MENA countries may lead to faulty estimation of risk in the world of volatile markets.Originality/valueThe paper tried to fill the gap in the literature and perform an evaluation of the relative performance of the most popular VaR estimates with an emphasis on the EVT methodology in seven MENA emerging stock markets. A comparison of the performance between EVT and other VaR techniques should support the decision whether more or less sophisticated methods are appropriate in order to assess stock market risks in the MENA countries.

Extreme observations and risk assessment in the equity markets of MENA region: Tail measures and Value-at-Risk

“…The EVT techniques make it possible to concentrate on the behavior of extreme observations in these markets. Consistent with the results from other developing countries (see Gencay & Selcuk, 2004;Maghyereh and Al-Zoubi, 2006), for all markets, it is shown that returns distributions are fait-tailed, and the return generating process lay in the domain of attraction of a Frechet distribution. We show that the Hill estimates of the tail indices are of roughly similar magnitude across all markets, and then estimate the market risk in for each country by measuring the average time before a given extreme value occurs and the corresponding Value-at-Risk in each market.…”

Measuring Extreme Market Risk: The Sri Lankan Context

“…Therefore, the method of choice for the ASPI is definitely the historical simulation method, as it performs well when backtested and is also very simple to implement. These findings are contrary to other empirical studies done on other stock indices (Harmantzis, Miao, and Chien 2006;Maghyereh and Al-Zoubi 2006). Comparing the dynamic models, we find that both the Two-Step Approach and the ARMA(0,1) GARCH(1,1)-n performed well at all three tests, but the Two-…”

“…The finding that the historical simulation method is the most suitable unconditional model is contrary to other empirical studies done on other stock indices (Harmantzis, Miao, and Chien 2006;Maghyereh and Al-Zoubi 2006). This signifies that, for the ASPI, performance in the past gives a reasonable idea of the performance in the future.…”

Extreme value theory in emerging markets: Evidence from the Montenegrin stock exchange

“…Bao, Lee, and Saltoglu (2006) examined the stock markets of five Asian economies (Indonesia, Malaysia, Korea, Taiwan, Thailand) during the financial crisis and showed that some EVT-based models did better in crisis periods. Maghyereh and Al-Zoubi (2006) showed that the EVT approach is superior to classical variance-covariance and historical simulation for seven Middle Eastern and North African countries. Cotter (2004Cotter ( , 2007 found that EVT estimates for Value at Risk and expected shortfall outperform other estimates, analysing six Asian markets during the Asian crisis and five equity indices from European markets.…”