A FORECASTING APPROACH TO REAL EFFECTIVE EXCHANGE RATE-OIL PRICE NEXUS IN CHINA

Abstract

Motivated by the theoretical link between real exchange rates and oil prices, we utilize a univariate moving average (MA) and an augmented MA (A-MA) model to generate multi-period forecasts of China’s real effective exchange rate for 2008–2018. The MA model utilizes past information in real exchange rates, and the A-MA model utilizes past information in both real exchange rates and oil prices. We show that the A-MA forecasts are unbiased and embody useful predictive information beyond that contained in the MA forecasts. In addition, the A-MA forecasts are directionally accurate under asymmetric loss. Such accurate forecasts are useful as inputs for policymakers to design an optimal real exchange rate policy to promote trade and attract foreign investment, and for foreign entities that regard China as an attractive environment for investing in various sectors.