Non-Linear Exchange Rate Relationships: An Automated Model Selection Approach with Indicator Saturation

This paper examines whether the explanatory power of exchange rate models can be improved by allowing for cross-country asymmetries and non-linear effects of fundamentals. Both appear to be crucial. The data set looks at the USD versus pound and yen exchange rates from 1982:07-2013:10, and bias-corrected automated model selection is conducted with indicator saturation. Several non-linear effects are significant at the 1% level and reveal that larger changes in fundamentals typically generate changes in the exchange rate at an increasing rate, with the exception of variables eliciting countervailing central bank reactions (for example consumer prices) which exhibit diminishing effects.