Estimation of value at risk and analysis of the presence of long memory on the indices of the selected emerging financial markets

---

Thesis Type: Postgraduate

Institution Of The Thesis: Dokuz Eylül University, Sosyal Bilimler Enstitüsü, İktisat Ana Bilim Dalı, Turkey

Approval Date: 2016

Thesis Language: Turkish

Student: FREDYNANDY MATOKE JOHN

Supervisor: HAKAN KAHYAOĞLU

Open Archive Collection: AVESIS Open Access Collection

Abstract:

This study was conducted with two main goals in mind. The first goal was to understand the impact of long memory in the return of the indices of emerging markets (Brazil, Russia, India, China and Tanzania). The second goal was to understand the potential risk of loss Value at Risk (VaR) if one is to invest in these financial markets. Our study employs ARFIMA – FIEGARCH model because unlike FIGARCH model, FIEGARCH model captures and better explains asymmetric volatility. The findings of this study shows that four markets' indices namely IBOVESPA, RTS, CNX 100 and SSE are characterized by asymmetric volatility indicated by negative coefficient of the leverage estimates. DSE does not indicate negative asymmetric volatility effects implying that negative and positive shocks of the same magnitude may have similar impact on its return volatility. Moreover, our study shows DSE has the lowest VaR. This can be explained by the least correlation DSE might have with the developed financial markets. Furthermore, study findings shows the presence of ARCH and GARCH effects in these markets at different level and magnitude and a clear evidence of the presence of long memory, implying that historical prices of these indices can be used to predict the future outcome of these financial markets.