A wavelet transform analysis of the relationship between unexpected macroeconomic news and foreign exchange rates

Abstract

This study applies the wavelet transform technique to analyse the relationship between exchange rates of major currencies and the unexpected macroeconomic news. Hourly closing exchange rates of major currencies from 2000 to 2009 are utilized for wavelet analysis and those of 2010 are used for ex ante test. It is found that JPY and CHF are the most affected exchange rates, which are followed by EUR and GBP. The least affected exchange rates are AUD and CAD. In general, the most influential news is NFP, which is followed by ISM_M. These two factors have an impact on at least four currency exchange rates in the short, intermediate and long terms. UR is a short-term factor and the effect of NFP lasts for all the three time horizons considered. Ex ante test is carried out and the accuracy rates for EUR, AUD, GBP, CAD, CHF and JPY are 0.75, 0.5, 0.67, 0.83, 0.83 and 0.58, respectively.