Arifudin Primary Author mixed material bibliography UNIVERSITAS HALU OLEO/ KENDARI FITK/Teknik Pertambangan 2018 monographic id Indonesia

Text

xi, 64 hal,; 28 cm ABSTRACT ANALYSIS OF IRON ORE PRICE PREDICTION USING TIME SERIES METHOD Arifudin Departement of Mining Engineering, Faculty Of Earth Science and Technology, Halu Oleo University, Kendari arif.bashar07@gmail.com The quantity of Indonesian imports of iron ore in 2012 to presently is increase. To prevent the amount of import costs, a predisction of iron ore price is needed, so that it can be use as a reference in importing. The method that can be use for prediction is the time series method. This method has several calculation methods, that is single moving average method, double moving average, weight moving average, trend projection, single exponential smoothing, double exponential smoothing one parameter from brown, double exponential smoothing two parameters from holt, holt winter additive model, holt winter multiplicative model. The purpose of this study is to determine the value of iron ore price predictions in 2018, 2019, 2020, 2021 and 2022, and determine which method is suitable for use in predictions, and has the highest level of accuracy. The data used is the price of iron ore year on year in 1971 to 2017. The results of this study indicate that the appropriate method for predicting iron ore prices that have a trend pattern and has the highest accuracy is double exponential smoothing two parameter method from holt, with the results of iron ore price prediction in 2018 is 70.13 $/dmtu, in 2019 is 70.83 $/dmtu, in 2020 is 71.54 $/dmtu, in 2021 is 72.24 $/dmtu, in 2022 is 72.94 $/dmtu, with a MAPE value of 16%, and a good predictive ability category. Key words : Iron Ore, Prediction Price. Teknik Pertambangan UPA PERPUSTAKAAN UHO UNIVERSITAS HALU OLEO F1B313118 UPT PERPUSTAKAAN UNIVERSITAS HALU OLEO F1B313118 34664 2019-01-29 10:19:11 2019-01-29 10:20:13 machine generated