On Some Deep Learning Algorithms and Grey Models for Forecasting FX Rates: A Comparative Study.

Volatility in the foreign exchange market is often associated with risk and may hence cause a stir in the sustainable development of an economy. The Mauritian economy, being open and globally integrated, is highly sensitive to fluctuations in other currencies. Consequently, volatility modelling and forecasting of the FX market is crucial for proper risk management exercises. This paper provides a comparative study on the FX rate modelling and forecasting accuracy of some conventionally used deep learning approaches against some grey models. In particular, we employ the recurrent neural networks and the long short-term memory recurrent networks using high-frequency historical data against the optimised Fourier grey Markov model (FOGM) using a significantly smaller dataset. We observe that the FOGM model with a smaller dataset outperforms the deep learning approaches considered.