CMStatistics 2016: Start Registration
View Submission - CFE
Title: Intra-day dynamics of US-dollar exchange-rates: Evidence from quantile autoregressions Authors:  Konstantin Kuck - University of Hohenheim (Germany) [presenting]
Robert Maderitsch - University of Hohenheim (Germany)
Abstract: The aim is to test several hypotheses regarding intra-day dynamics in exchange rates. Specifically, we investigate the presence of intra daily ``over''- and ``undershooting'' using quantile autoregressions based on intra-day returns over time-horizons from 5-min up to several hours. The exchange rates that we consider are the USD/EUR, USD/GBP and USD/JPY. Our data-set covers an 11-year period from January 03, 2000 to September 30, 2011, which enables us to provide a long-term perspective on intra-daily dynamics and to investigate potentially changing properties in phases of increased uncertainty in financial markets. In line with previous studies, we find that the intra-daily return autocorrelation is mostly negative which is typically associated with market microstructure effects. In the tails of the return distribution, however, a tendency for positive but insignificant dependence can be observed. We moreover document that temporal dependence declines when intra-day volatility rises due to trading activity. Finally, the dynamics appear to be remarkably stable across different market conditions, that is, they seem not to be affected by increased uncertainty in the financial markets. Overall, our findings show that both over- and undershooting are absent in US-Dollar exchange rates and indicate that information are processed efficiently in the sense that future returns cannot be predicted based on historical returns.