Peramalan Nilai Inflasi di Indonesia Menggunakan Double Exponential Smoothing dan Triple Exponential Smoothing

Abstract

Inflation is a situation where there is a tendency for the price of goods to increase in general over a long period of time in a country. Monitoring the value of future inflation is very important to know, because the inflation rate has a significant impact on economic growth. Therefore, accurate and fast forecasting is needed. There are several methods that can be used to forecast, including double exponential smoothing and triple exponential smoothing. Parameters with the double and triple exponential smoothing methods often experience exponential increases or decreases in predicted values. These values ​​are often much larger or smaller than the actual data. To overcome this, a parameter is added that can dampen exponential growth, namely using the damped parameter. The damped parameter value is added to each trend smoothing. This study uses inflation data in Indonesia from 2017-2023. The purpose of this study is to determine the results of forecasting inflation values ​​in Indonesia using double exponential smoothing and triple exponential smoothing with or without damped parameters and to determine the best method. The results of this study are that the double exponential smoothing parameter damped method is very good to use based on a comparison of the smallest MAPE value of all the methods used with a MAPE value of 9.63% and the forecast results for January 2024 of 0.0261