On the efficiency of the gold market: Results of a real-time forecasting approach

Abstract

Using a real-time forecasting approach, we study whether publicly available information on a large set of financial and macroeconomic variables help in forecasting out-of-sample monthly excess returns on investing in gold. The real-time forecasting approach accounts for the fact that an investor must reach an investment decision in real time under uncertainty concerning the optimal forecasting model. The real-time forecasting approach also accounts for the possibility that the optimal forecasting model may change over time. We account for transaction costs and show that using forecasts implied by the real-time forecasting approach to set up a simple trading rule does not necessarily lead to a superior performance relative to a buy-and-hold strategy, implying that the gold market is informationally efficient with respect to the predictor variables that we study in this research.