Zarmina Ali Khan* and Arshad Hassan
Purpose: This study analyse the asymptotic behavior of the tails of the return distributions in emerging markets through the lens of extreme value theory. It estimates and compares efficacy of the EVT based value at risk in emerging markets like Brazil(Bo Vespa), Russia(MOEX), India(Nifty 50), Bahrain(Share BAX), China(Shanghai), Colombia(COLCAP), Malaysia(FTSE), Thailand(SET INDEX), Argentina(Marvel), Bangladesh(Dhaka Stock Exchange), Pakistan(KSE 100) and Sri Lanka(CSE)) by using daily data for the period 2000-2018.
Methodology: The study employs block maxima model (BMM) based on generalized extreme value distribution (GEV) and peak over threshold model (POTM) based on generalized Pareto distribution (GPD). Peak over threshold model is applied under the assumption of unconditional and conditional volatility. Use of conditional EVT to estimate VaR is proposed by McNeil and Frey (2000) to capture the heteroscedasticity in extreme returns arising from stochastic volatility. Finally, the efficacy of models is evaluated through backtesting techniques proposed by Kupiec (1995) and Christoffersen (1998).
Findings: The Block maxima model underestimates risk whereas condition POT model overestimates risk at 95% and 99% confidence level. EVT based unconditional POT model performs better in comparison to the GEV based block maxima model and EVT based conditional POT model in all selected countries of emerging markets. The results are consistent under various backtesting approaches as the performance of the model does not change with an increase in confidence level. The difference of risk may be the outcome of role of equity markets in a country, levels of disclosure, governance laws and ownership structure.
Value: This study has important implication for portfolio managers in making the decision regarding resource allocation, portfolio diversification and risk management as EVT based unconditional POT model can be used for modeling risk of extreme events in emerging markets.
HTML PDFShare this article
