Abstract:
This study aims to capture volatility patterns using GARCH (1,1) models.
It evaluates these models to obtain one-step-ahead forecastabilities by employing
four major forecasting evaluation criteria, and compares two different currencies—
the Pakistan rupee and the US dollar—as domestic and foreign currency-valued
exchange rates, respectively. The results show that using an international vehicle
currency is favorable in Pakistan’s context. However, the Kuwaiti dinar, Canadian
dollar, US dollar, Singapore dollar, Hong Kong dollar, and Malaysian ringgit are
found to be preferable when performing direct international transactions. Using
the root mean square errors and mean absolute errors techniques, the study also
assess the robustness of measuring one-step-ahead forecasts.
