A COMPARATIVE-ANALYSIS OF GOLD MARKET-EFFICIENCY USING DERIVATIVE MARKET-INFORMATION

The concept of market efficiency is the foundation for several theoretical constructs in financial economics. This paper tests for market efficiency when information from one market is used to predict prices in a different market. We examine both time-series and multiple regression models and compare the power of each of these models from both a statistical and economic viewpoint. Significant correlations between the gold spot market price and other market variables suggest that market inefficiencies exist. For our particular sample and tests, the time-series model is much superior to the multiple regression model. However, the results also suggest that a transfer function model which is a weighted average of the two models could be an effective forecasting model.