IMPLEMENTATION OF THE DOUBLE EXPONENTIAL SMOOTHING (DES) METHOD TO FORECET THE CONSUMER PRICE INDEX (CPI) IN MEDAN CITY

Abstract

The Central Statistics Agency (BPS) as a non-ministerial government agency has the task of providing data needs for the government and the public obtained through censuses and surveys conducted by BPS itself and also from other government departments or agencies as secondary data. The problem that occurs at BPS is the field of inflation, one of which is in the field of the consumer price index, which experiences price fluctuations which have an impact on the economy. Forecasting is an important tool used for effective and efficient planning, therefore, forecasting is needed to predict various events that will occur in the future. One of the methods used for forecasting is the linear Double Exponential Smoothing (DES) method from Brown, this method basically uses past data which is smoothed by exponen- tially weighting the older observation values or newer values and the data used shows trend pattern. Trend is a smoothed estimate of the average growth at the end of the period. In this study, the best parameter α for forecasting the consumer price index in the city of Medan is α= 0,8 with a MAPE percentage of 0,0223. And the results of the consumer price index forecast in the city of Medan in 2022-2023 show an increase every month.