Fluctuating gold prices can result in losses in investments, thus it needs forecasting to reduce the risk of loss in making gold investment decisions. This research builds the application of gold price forecasting using moving average method. Moving average is a method of forecasting with the price of gold in the past. In this study, the moving averages used two different parameters, if the moving average value was above the gold price, it is likely the price will drop. If the moving average value is below the gold price, it is likely that the price will rise. The program implementation uses the Meta Quotes Language (MQL) programming language, which is located on Meta Trader 4 (MT4). System testing is done using backtest simulation on MT4. System test results obtained 99.85% accuracy level with absolute mean error (MAE) Rp. 7,672 or 1.56%. The results of system interface testing conducted using questionnaires acquired usability factors of 3.96, simplicity factor of 3.78 and interactivity factor of 3.78 from a scale of 1 to 5. From the first and second test results show that the system has been well implemented and easy to use.

Keywords : Backtest Simulation, Gold, Moving Average, MQL, Forecasting Efisiensi.