Abstract:
The decision of the United Kingdom to leave the European Union (Brexit) after 43 years
caused turmoil in exchange rate and global stock markets. More specifically, the pound
relative to the dollar has lost close to 15 percent of its value in the weeks after the
Brexit decision. In this paper we attempt to examine whether this sudden depreciation
of the (pound-dollar) exchange rate is the reaction of market participants to the Brexit or
whether the exodus of UK from the EU had little impact on the exchange rate. In doing
so, we train linear and nonlinear econometric and machine learning models and evaluate
out-of-sample forecasts of the exchange rate and its realized volatility in the pre- and
post-Brexit period. We quantify uncertainty caused by the Brexit according to an index
based on news related to economic uncertainty. We argue that in daily forecasting horizon
our models adhere closely to the evolution of the exchange rate and that most of the
depreciation is based on the uncertainty caused by the Brexit
