ENSEMBLE KALMAN FILTER PADA PEMULUSAN GANDA DUA PARAMETER DARI HOLT

Abstract

ABSTRACT

Forecasting will always produce errors as well as the model or forecasting method used, therefore it is very important to reduce these errors or errors. This study aims at applying the Ensemble Kalman Filter method to the forecasting model by smoothing two parameters from Holt and obtaining forecasting results with a smaller percentage of error than the results of trial and error. The Ensemble Kalman Filter method is offered as a way to reduce errors generated by Holt's Two-Parameter Double Exponential Smoothing model. In the study in this paper, it is proven that the combination of the Ensemble Kalman Filter on Holt's Two Parameter Exponential Smoothing can reduce errors significantly.