GOLD PRICE PREDICTION BASED ON OIL PRICE FLUCTUATIONS USING TIME SERIES FORECASTING MODEL ARIMA AND LSTM

RAHARDJO, JULIO IRVAN (2024) GOLD PRICE PREDICTION BASED ON OIL PRICE FLUCTUATIONS USING TIME SERIES FORECASTING MODEL ARIMA AND LSTM. Skripsi thesis, UNIVERSITAS KATOLIK SOEGIJAPRANATA.

[img] Text
20.K1.0004-JULIO IRVAN RAHARDJO_COVER_1.pdf
Download (283kB)
[img] Text
20.K1.0004-JULIO IRVAN RAHARDJO_BAB I_1.pdf
Restricted to Registered users only
Download (173kB)
[img] Text
20.K1.0004-JULIO IRVAN RAHARDJO_BAB II_1.pdf
Restricted to Registered users only
Download (183kB)
[img] Text
20.K1.0004-JULIO IRVAN RAHARDJO_BAB III_1.pdf
Restricted to Registered users only
Download (607kB)
[img] Text
20.K1.0004-JULIO IRVAN RAHARDJO_BAB IV_1.pdf
Restricted to Registered users only
Download (1MB)
[img] Text
20.K1.0004-JULIO IRVAN RAHARDJO_BAB V_1.pdf
Restricted to Registered users only
Download (167kB)
[img] Text
20.K1.0004-JULIO IRVAN RAHARDJO_DAPUS_1.pdf
Download (288kB)
[img] Text
20.K1.0004-JULIO IRVAN RAHARDJO_LAMPIRAN_1.pdf
Restricted to Registered users only
Download (294kB)

Abstract

Gold and oil prices are commodities that influence and have a significant impact on the global economy. This research aims to analyze the relationship between gold and oil prices and incorporate them into ARIMA and LSTM algorithms that will be compared to find an algorithm that can predict future gold prices more accurately. Previous research has shown that rising real oil prices have a positive effect on gold prices. Taking this relationship into account, accurate gold price predictions can help market participants to make informed investment decisions. The limitation of this study is that it only focuses on oil price fluctuations and does not include other factors that affect the gold market. The goal is to provide accurate gold price predictions for the next one year. The best prediction algorithm is LSTM which produces Mean Squared Error 2488.0454, Mean Absolute Error 37.7266, Mean Absolute Percentage Error 2.52%. Gold price predictions for the period November 2023 to October 2024 range from 1659.7815 - 1884.176US$. The findings contribute to the development of reliable gold price predictions based on oil price fluctuations and help in making the right investment decisions for stable economic growth.

Item Type: Thesis (Skripsi)
Subjects: 000 Computer Science, Information and General Works
Divisions: Faculty of Computer Science > Department of Informatics Engineering
Depositing User: Mr Yosua Norman Rumondor
Date Deposited: 17 Apr 2024 07:11
Last Modified: 17 Apr 2024 07:11
URI: http://repository.unika.ac.id/id/eprint/35192

Actions (login required)

View Item View Item
Unika Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.