Abstract
This study empirically analyses exchange rate risk in a portfolio of ten stock indices in emerging markets from the viewpoint of a South African investor. The aim of this study is to understand the effect of exchange rate risk on expected return of such a portfolio and to find out whether this investment provides benefits for South African investors. To this end, we covered ten stock markets namely Malaysia, Philippines, South Africa, Brazil, China, Russia, India, Argentina, Mexico and Singapore. We make use of Value at Risk (VaR)-based GARCH model to model extreme currency deviation.
We collect daily stock prices, daily spot and forward exchange rate of the South African rand against the currencies of the above mentioned countries for the period between 1 January 2005 and 31 October 2016. Firstly we filter each market returns with an Exponential Generalised Autoregressive Conditional Heteroscedasticity (EGARCH) model to eliminate the presence of heterokedasticity and autocorrelation in the distribution of returns. We then fit the residual of the above GARCH model to the Generalized Pareto distribution (GPD) in order to account for extreme events.
The estimation results of the EGARCH (1, 1) model show that all parameters are statistically significant for all stock markets including the leverage effect which arguably proves that bad news have higher impact on stock market volatility compared to good news. The parameters of the GPD in the lower tail are also estimated, the resulting shape parameters are significantly positive indicating that these stock markets are prone to price swing during periods of economic downturn.
Lastly, we compute individual market risk measures using the EGARCH-EVT – based techniques and backtest them using the Kupiec method. Our VaR show that Russia has the highest market risk whereas Malaysia has the lowest VaR implies that it is the lowest market risk. The result of unconditional coverage test show that the likelihood ratio statistic fails to reject the null hypothesis of correct number of exceptions implies that our model is accurate. We find that for a South African investor to maximize his/her entire portfolio return, S/he needs to invest 9.85% in ALSI, 6.59% in SHANGHAI, 4.44% in BOVESPA, 7.46% in SENSEX, 2.28% in MICEX, 12.15% in MEXBOL, 3.9% in MERVAL, 31.4% in KLCI, 12.74% in SINGA and 9.19% in PSEI with the portfolio risk of 1.59; 0.62; 1.3; 5.76; 2.48; 1.45; 1.36; 0.24 and 2.21 respectively. We analyses the impact of exchange rate risk on the portfolio, the result show that Singaporean dollar, Russian rouble, Mexican peso and Indian rupee have positive impact on the portfolio return while the Argentine peso and Chinese yuan have negative impact on portfolio return therefore, South African investor in order to maximize his investment taking into account exchange rate risk have to put more weight in stock market with positive impact of exchange rate fluctuation on the portfolio.
M.Com. (Financial Economics)