Explaining the dynamics of exchange rate volatility

Sammanfattning: This research examines the volatility of the Swedish krona in regards to the Euro and US-dollar exchange rate, using both daily and monthly data ranging from the beginning of 2000 until 2022. Using this time span allows us to update previous literature on exchange rate volatility, and also incorporates recent economic events such as the great financial crisis of 2008, the 2020 covid-pandemic and the geopolitical uncertainty in Europe following Russia's invasion of Ukraine. In order to model the volatility of the exchange rate, Generalized Auto Regressive Conditional Heteroskedasticity (GARCH) models are used in order to account for dependencies in the error term. The research also incorporates Exponential GARCH (EGARCH) models that allow for the effect of positive and negative developments in the return of the exchange rate impacting volatility differently (Positive price shocks affect the volatility differently compared to negative). We extend the ARCH and GARCH models with explanatory variables that are grounded in previous literature as well as economic theory. The daily variables used in this paper are the volatility index, financial condition index, a proxy for hedge-fund positioning, a constructed news proxy and also controlled for the rate of return of the EUR/USD exchange rate. Over a monthly basis, we study the monetary aggregates of the US, Eurozone, and Sweden in addition to inflation and the Bloomberg economic sentiment index. This paper finds that there are several ARCH, GARCH, and E-GARCH effects present in the exchange rate volatility for both EUR/SEK and USD/SEK for both frequencies, suggesting that negative and positive price developments affect the volatility differently and that the exchange rates exhibit conditional heteroskedasticity. In terms of explanatory long-run variables we find a significant positive relationship between monthly US monetary policy and a proxy for overall market sentiment to affect exchange rate volatility. For the daily frequency, this paper finds a strong negative relationship between EUR/USD return and volatility for both EUR/SEK and USD/SEK.