Forecasting foreign exchange rates with regression networks tuned by Bayesian optimization
This work addresses forecasting accuracy and interpretability for financial analysts, though it is incremental as it builds on existing regression and optimization methods.
The paper tackles multi-step forecasting of foreign exchange rates by introducing RegPred Net, a regression network that models rates as a generalized Ornstein-Uhlenbeck process and tunes hyperparameters with Bayesian optimization. It shows significant improvements over models like ARIMA and LSTMs on metrics such as RMSE and R-squared for rates like EUR/USD over a 100-day horizon.
The article is concerned with the problem of multi-step financial time series forecasting of Foreign Exchange (FX) rates. To address this problem, we introduce a regression network termed RegPred Net. The exchange rate to forecast is treated as a stochastic process. It is assumed to follow a generalization of Brownian motion and the mean-reverting process referred to as the generalized Ornstein-Uhlenbeck (OU) process, with time-dependent coefficients. Using past observed values of the input time series, these coefficients can be regressed online by the cells of the first half of the network (Reg). The regressed coefficients depend only on - but are very sensitive to - a small number of hyperparameters required to be set by a global optimization procedure for which, Bayesian optimization is an adequate heuristic. Thanks to its multi-layered architecture, the second half of the regression network (Pred) can project time-dependent values for the OU process coefficients and generate realistic trajectories of the time series. Predictions can be easily derived in the form of expected values estimated by averaging values obtained by Monte Carlo simulation. The forecasting accuracy on a 100 days horizon is evaluated for several of the most important FX rates such as EUR/USD, EUR/CNY, and EUR/GBP. Our experimental results show that the RegPred Net significantly outperforms ARMA, ARIMA, LSTMs, and Autoencoder-LSTM models in terms of metrics measuring the absolute error (RMSE) and correlation between predicted and actual values (Pearson R, R-squared, MDA). Compared to black-box deep learning models such as LSTM, RegPred Net has better interpretability, simpler structure, and fewer parameters.