Abstract
Foreign exchange rate movements have increasingly become a concern for practitioners, poli-cymakers, and researchers due to the volume of transactions in foreign exchange rate markets and the stylised features of the assets, including volatility spillover, volatility clustering, and heavy-tailed volatility. These features have a large impact on portfolio allocation and macroe-conomic stability. Therefore, it is of great importance to analyse the foreign exchange move-ments and the risk and dependence structures to subsequently propose an efficient foreign ex-change rate policy and risk management strategy. Doing so is important because foreign ex-change rate volatility has resulted in the associated risk being considered one of the major issues with which multinational firms and fund managers contend in international portfolio diversifi-cation. Moreover, economic activities such as importing and exporting have motivated research interest, with modelling and forecasting exchange rate volatility becoming global concerns for investors. Many financial researchers, both practitioners and those in academia, have thus in-creasingly shown interest in measuring and forecasting exchange rate volatility and examining the behaviour of financial markets.
Controlling for currency fluctuations past through understanding the dependence structure, which is very significant for diversification purposes and enables decisions makers to under-stand how one market responds to currency fluctuations in other markets. Fund managers and investors need to pay attention to the dependence structures between exchange rate volatility and equity markets in international investment decision-making. Therefore, this thesis explores the dependence pattern between foreign exchange rate movement and some macroeconomic factors to quantify the risk associated with foreign exchange rate exposure and predict foreign exchange rate movement direction. To do so, we have conducted three empirical analyses fo-cused on modelling those movements.
The first empirical chapter of this thesis investigates the linkages between macroeconomic var-iables and foreign exchange rate direction and predicts the latter using an ensemble method. We classified exchange rate movements into two classes, appreciation and depreciation. We employed monthly data for macroeconomic variables and the currency price of the US dollar, Euro, Japanese yen, and British pound against the South African rand. The results showed that the ensemble method provides an accurate prediction for the appreciation of the euro, US dollar, and British pound and depreciation of the Japanese yen. Additionally, we found that the stock
vi
price and terms of trade are more responsible than the other variables for the appreciation of the US dollar.
Chapter three focuses on quantifying foreign exchange rate risk and modelling the dependence structure of underlying assets with selected listed stock indices. We proposed a robust risk model based on stochastic volatility- extreme value theory ( SV-EVT)-pairwise copulas. Our empirical findings show that the Glosten–Jagannathan–Runkle- Generalised autoregressive conditional heteroscedasticity (GJR-GARCH) with Student’s t-distribution combined with a regular-vine (R-vine) copula outperforms the alternative models. Dependence structure analysis revealed a strong co-dependency between the stocks from the financial industry and foreign exchange rates; we also found that the value at risk (VaR)-based R-vine copula model outper-formed the VaR-based D-vine and C-vine models.
The last empirical chapter focuses on modelling foreign exchange rate co-movement and its spatial dependence in emerging markets. We employed the spatial Durbin model to investigate the extent to which macroeconomic factors influence the co-movement of foreign exchange rate markets. Our empirical findings showed that not only do the macroeconomic factors influ-ence foreign exchange rate co-movement but cultural proximity to other emerging markets. We also found that terms of trade, inflation differential, and remittance are the most important fac-tors affecting foreign exchange rate movement.