Abstract
In this article we use a generalization of the standard nearest neighbours, called local regression (LR), to study the predictability of the yen/US$ and pound sterling/US$ exchange rates. We also compare our results with those previously obtained with global methods such as neural networks, genetic programming, data fusion and evolutionary neural networks. We want to verify if we can generalize to the exchange rate forecasting problem the belief that local methods beat global ones.
Acknowledgements
Marcos Álvarez-Díaz gratefully acknowledges Ministerio de Educación y Ciencia (Grant MTM2005-01274, FEDER funding included) for its financial support, and Pacific Exchange Rate Service for providing the data.