MODEL FOR FORECASTING THE TREND IN THE SALE PRICES OF DAILY GOLD BAR

Abstract

The objectives of this study were to develop a model for forecasting the daily sale prices of gold bars, to compare the accuracy and suitability of two forecasting methods, and to use the chosen model to forecast gold bar sale prices from August 1, 2022, to December 31, 2022. The methods applied in this study were Brown’s Exponential Smoothing Method and Holt’s Exponential Smoothing Method, using daily time series data spanning from January 1, 2018, to July 31, 2022, totaling 1,673 days. This dataset was divided into two subsets. The first subset, covering January 1, 2018, to June 30, 2022 (1, 1,642 days), was used to construct the forecasting models. The second subset, comprising data from July 1, 2022, to July 31, 2022 (31 days), was used to evaluate and compare the models’ performance. Model accuracy was assessed using the Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE). The results indicated that Holt’s Exponential Smoothing Method outperformed Brown’s Method, achieving a lower MAPE of 0.86 and a lower RMSE of 295.66. Based on this performance, Holt’s method was selected as the preferred model for forecasting future gold bar sale prices. The forecasting results revealed a downward trend in gold bar sale prices during the prediction period, suggesting that prices were likely to decline through the end of 2022.