<![CDATA[Rice is a staple food for Indonesian people whose price often fluctuates. Throughout the period from January 2021 to September 2024, the price of rice in East Kalimantan continued to increase, forming a trend pattern. Data forecasting with a trend pattern can be done using the Brown's Weighted Exponential Moving Average (B-WEMA) method which uses smoothing, weighting, and order parameters in its calculations. In this method, a larger weighting is given to the latest data and the order used is a high order. In addition, determining the smoothing parameters that provide the best accuracy value is usually done by the trial-and-error method. To increase the accuracy value of the B-WEMA method, smoothing parameter optimization can be carried out using the Levenberg-Marquardt method. The purpose of this study was to apply the B-WEMA method with Levenberg-Marquardt optimization to rice price forecasting in East Kalimantan. The results showed that the smoothing parameters obtained by the B-WEMA method before optimization were 0,7 with an accurate value of MAPE is 0,820%. After optimization, the optimal smoothing parameter obtained was 0,652 which resulted in an accurate value of MAPE is 0,818%. This indicates an increase in the accuracy value. Based on the optimization results, the forecast results for the next 3 periods were obtained, namely Rp16.070 for October; Rp16.027 for November; and Rp15.985 for December.]]>
Copyrights © 2025
