Modeling exchange rate dynamics: Exchange rates as a function of fundamentals, market herding, and Central Bank intervention
This dissertation proposes a model for forecasting spot exchange rates and exchange rate volatility. The model, which posits that exchange rate dynamics is determined by three groups of variables, namely, macroeconomic fundamentals, market herding, and the impact of reported or rumored Central Bank (CB) intervention, was tested using the USD-Euro exchange rate for the period January 1999 to December 2009. Results of the empirical tests show that the model performed well in forecasting daily spot rates and monthly FX volatilities. In daily spot rates, the model was slightly less efficient than random walk but more efficient than Dornbusch-Frankel. For the volatility forecast, the model was slightly more efficient than the market. A possible explanation for the model's performance might be found in Frankel (1991) who noted that participants could improve their volatility forecasts by putting more weight on the long-run average. By incorporating variables for macro-economic fundamentals, market herding behavior, and impact of CB intervention, the model provided a method for estimating the weight that should be assigned for the fundamentals as well as for the extrapolating or trend-chasing tendencies of the market. ^
Austria, Carolina Pondang, "Modeling exchange rate dynamics: Exchange rates as a function of fundamentals, market herding, and Central Bank intervention" (2012). ETD Collection for Fordham University. AAI3512217.