Predicting Returns of Exchange Rate from Oil Prices: Machine Learning Approach

Abstract

Using data collected monthly beginning in January 2010 and ending in December 2022, this research intends to forecast the returns of the Vietnam Dong/US Dollar Exchange Rate on the oil price worldwide. Following the current literature, this investigation develops a predictive model that considers the most important aspects of the predictor and the predicted series. Oil and the returns on the exchange rate were shown to have a linearly negative connection. Both net oil exporters and net oil importers might anticipate favorable exchange rate returns if oil prices remain stable. Finally, we evaluate how different forecasting methods perform in and out of the sample. While there was higher volatility in the out-of-sample period, the research indicated that the prediction error was still small.