Essays in the study and modelling of exchange rate volatility

The thesis is a contribution to the literature on the study and modelling of exchange rate variability, and contains seven chapters. Chapter 1 motivates the study by providing a historical context about its importance, sketches the main themes of the thesis, and gives an overview of the Norwegian economy since the empirical studies are on the Norwegian krone against the Euro exchange rate. Chapter 2 makes a distinction between period and within-period exchange rate variability, a distinction which is of special interest when studying variability across different exchange rate regimes. Also, the exponential model of variability (EMOV) is put forward as a particularly convenient framework for explanatory exchange rate variability modelling. Chapter 3 makes full fuse of these ideas in studying the impact of market activity on exchange rate variability in the case of Norway. The main findings of this study are that the impact of short-term change in market activity, as measured by relative week-to-week changes in quoting frequency, is positive and statistically significant for both definitions of variability, and that the impact is relatively stable across three different exchange rate regimes. Also, our results do not suggest that the persistence in variability can be explained by persistence in the level of quoting. Chapter 4 undertakes an out-of-sample forecast evaluation of general to specific (GETS) modelling of exchange rate volatility, and our results suggest GETS-derived models are particularly useful in conditional forecasting. Chapter 5 studies the relation between exchange rate variability, market activity and heterogeneity using a measure of spot NOK/EUR transaction volume from banks within Norway's regulatory borders. Our results do not support the hypothesis that short-term Norwegian market activity has an impact on variability. However, we do find some support of the hypothesis that large and small banks have an opposite impact through their long-term activity. Chapter 6 proposes a solution to a shortcoming in the first stage of David F. Hendry's reduction theory by interpreting the underlying outcome space as a set of possible worlds made up of indeterministic and historically inherited particulars. Finally, chapter 7 concludes and provides suggestions for further research.