Peramalan Harga Minyak Mentah Di Indonesia Dengan Menggunakan Metode Fuzzy Time Series

Widyandari L W, Trigati (2015) Peramalan Harga Minyak Mentah Di Indonesia Dengan Menggunakan Metode Fuzzy Time Series. Undergraduate thesis, Institut Technology Sepuluh Nopember.

[img] Text
5211100051-Undergradaute Thesis.pdf - Published Version

Download (3MB)

Abstract

Meramalkan pergerakan harga minyak sangat penting bagi pelaku bisnis dalam pasar energy. Hal ini dikarenakan minyak bumi merupakan input vital dalam proses produksi industri, terutama untuk menghasilkan listrik, menjalankan mesin produksi dan mengangkut hasil produksi ke pasar. Mengingat peranannya yang vital tersebut, maka dampak yang timbul akibat fluktuasi harga minyak akan berpengaruh terhadap bidang lain, diantaranya pertumbuhan ekonomi, laju inflasi, jumlah uang beredar, nilai tukar riil rupiah terhadap US dolar dan suku bunga.Data harga minyak mentah Indonesia merupakan data time series. Dikatakan time series dikarenakan data harga minyak mempunyai interval waktu yang sama dan diamati pada suatu periode tertentu. Selain itu data harga minyak memiliki subjektivitas terhadap pengkategorian harga minyak. Subjektivitas terhadap harga minyak merupakan nilai linguistik. Oleh sebab itu, dibutuhkan proses peramalan dengan mengkaitkan time series dan teori fuzzy. Sehingga pada penelitian ini akan dikembangkan metode Fuzzy Time Series Markov Chain Model untuk meramalkan harga minyak mentah Indonesia. Fuzzy time series menggabungkan sikap subjektif orang dan nilai-nilai obyektif historis dapat membantu memecahkan masalah peramalan. Dalam penelitian ini, data yang digunakan adalah data harian harga minyak mentah Indonesia dari tahun 2010 hingga 2014. Analisa model peramalan yang sesuai dengan harga minyak mentah Indonesia dimana selama ini selalu mengalami fluktuasi dilakukan dengan pengujian proporsi data, pengujian kinerja model, serta pengujian perbandingan peramalan dengan metode FTS Markov Chain dengan FTS biasa. Hasil pengujian menunjukkan bahwa model peramalan yang tepat untuk data harga minyak mentah Indonesia adalah dengan menggunakan metode peramalan Fuzzy Time Series Markov Chain Model, dengan proporsi data 70:30, yang menghasilkan eror MAPE SLC sebesar 2.35%, Arjuna 2.53%, Attaka 2.22%, Cinta 2.59%, Duri 2.44%, Widuri 2.56%, Belida 2.45%, dan Senipah Condensate sebesar 2.36%. ========================================================================================================= Forecasting the movement of oil prices is very important for businessmen in the energy market. This is because oil is a vital input in the process of industrial production, mainly to generate electricity, run the machine production and transporting produce to market. Considering the vital role these impacts caused by fluctuations in oil prices will affect other fields, such as economic growth, inflation rate, the amount of money in circulation, the real exchange rate of the rupiah against the US dollar and interest rates. Indonesia crude oil price data is the data time series. Time series is said to be due to the oil price data has the same time interval and observed in a given period. Besides oil price data have subjectivity against this oil prices. Subjectivity against oil price is the value of Linguistics. Therefore, it takes the process of forecasting time series and linked with the theory of fuzzy. So this will be developed on the research method of Fuzzy Time Series Markov Chain Model for predicting the price of crude oil Indonesia. Fuzzy time series combines the subjective attitude of the person and the objective historical values can help solve the problem of forecasting. In this study, the data used is the daily crude oil price data for Indonesia from 2010 to 2014. Analysis of forecasting model which corresponds to the price of crude oil which Indonesia has always been experiencing fluctuations in the testing done by the proportion of data, testing the performance of the viii model, as well as comparison testing of forecasting method with FTS Markov Chain with FTS. The test results show that the proper forecasting model for Indonesia crude oil price data is by using Fuzzy Time Series forecasting method of Markov Chain Model, with the proportion of 70:30 data, which produces error MAPE SLC of 2.35% 2.53%, Arjuna, Attaka 2.22%, love 2.59% 2.44%, thorns, Widuri 2.56%, Belida 2.45%, and Senipah Condensate of 2.36%.

Item Type: Thesis (Undergraduate)
Additional Information: RSSI 519.535 Wib p
Uncontrolled Keywords: Fuzzy Time Series, Harga Minyak Mentah Indonesia, Peramalan.
Subjects: Q Science > QA Mathematics > QA9.64 Fuzzy logic
T Technology > T Technology (General) > T174 Technological forecasting
Divisions: Faculty of Information Technology > Information System > (S1) Undergraduate Theses
Depositing User: Mr. Tondo Indra Nyata
Date Deposited: 14 Oct 2019 04:04
Last Modified: 14 Oct 2019 04:04
URI: http://repository.its.ac.id/id/eprint/71152

Actions (login required)

View Item View Item