Page 35 - Proceeding The 2nd International Seminar of Science and Technology : Accelerating Sustainable Innovation Towards Society 5.0

nd
The 2 International Seminar of Science and Technology
“Accelerating Sustainable innovation towards Society 5.0”
ISST 2022 FST UT 2022
Universitas Terbuka
1 INTRODUCTION
Forecasting is an activity with the aim of predicting what will happen
in the future [1][2]. There are several forecasting methods such as
Moving Average (MA), Exponential Smoothing (ES), Autoregressive
Integrated Moving Average (ARIMA) to Random Forest (RF). Single
Exponential Smoothing method is a method used in short-term
forecasting by assuming that the data fluctuates around a fixed
average value without a trend or consistent growth pattern [3].
Random forest was first introduced by Breiman in 2001, the Random
Forest method works by building a model using several decision trees
at random and combining the predictions of each tree to get prediction
results. Random Forest is a forecasting method that does not require
any assumptions [4].
This forecasting method can be applied in various fields, one of which
is the economic field, namely, to find out future gold demand. Gold is
a precious metal that is in high demand, both for investment and as
jewellery [5]. Gold has many enthusiasts because the value of gold
tends to be stable, easy to reach, easy to melt and of course promises
quite large profits, especially in Indonesia. Indonesia is one of the
countries in Southeast Asia that is included in the category of high
demand for gold. This is evidenced based on data from the World Gold
Council (WGC) that demand for gold jewellery in Indonesia in the
fourth quarter of 2015 experienced an annual growth of 16.88% from
7.7 tons to 9 tons. Throughout 2015, the demand figure reached 38.9
tons [6]. From 2016 to 2019, the demand for gold in Indonesian
jewellery continues to increase.
There are several previous studies that used the Random Forest and
Single Exponential Smoothing methods for forecasting, some of which
were carried out by Setyowati [7] namely comparing Exponential
Smoothing and Moving Average methods to predict motor vehicle
testing levies. The result of this study is that the Single Exponential
Smoothing method is better used in predicting the retribution for
testing motor vehicles with a MAPE value of 0.12%. Other research
was also conducted by Supriyanto [8] namely comparing the KNN,

14 ISST 2022 – FST Universitas Terbuka, Indonesia
International Seminar of Science and Technology “Accelerating Sustainable
Towards Society 5.0