Prediksi harga emas menggunakan support vector regression dan random forest regression = Gold price prediction using support vector regression and random forest regression

Satyawan, Elisa (2024) Prediksi harga emas menggunakan support vector regression dan random forest regression = Gold price prediction using support vector regression and random forest regression. Bachelor thesis, Universitas Pelita Harapan.

[img] Text (Title)
Title.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (250kB)
[img] Text (Abstract)
Abstract.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (594kB)
[img] Text (ToC)
ToC.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (584kB)
[img] Text (Chapter1)
Chapter1.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (658kB)
[img] Text (Chapter2)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (1MB)
[img] Text (Chapter3)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (818kB)
[img] Text (Chapter4)
Chapter4.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (2MB)
[img] Text (Chapter5)
Chapter5.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (570kB)
[img] Text (Bibliography)
Bibliography.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (579kB)
[img] Text (Appendices)
Appendices _watermark.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (4MB)

Abstract

Emas merupakan salah satu logam mulia yang aktif diperjual belikan oleh para investor dan manajer portofolio. Sebagai aset safe haven yang aman terhadap penurunan harga pasar, emas dapat melindungi nilainya, dapat digunakan untuk menstabilkan portofolio, dan menjadi sumber likuiditas dalam kondisi yang tidak stabil. Akan tetapi, emas memiliki nilai yang berfluktuasi dan terus mengalami perubahan dari waktu ke waktu. Oleh karena itu, memprediksi harga emas merupakan hal yang penting untuk meningkatkan keuntungan dan mengurangi kerugian. Prediksi dapat dilakukan dengan bantuan software RStudio dan menggunakan konsep data mining dalam bidang bisnis ataupun keuangan. Penelitian dilakukan menggunakan metode Support Vector Regression dan Random Forest Regression. Data yang digunakan adalah data harian dari emas (Gold Spot US Dollar), minyak (Bloomberg WTI Crude Oil), nilai tukar (US Dollar Index), dan saham (S&P 500) yang terdiri dari harga pembuka, penutup, tertinggi, dan terendah pada tahun 2020−2022. Penelitian ini diaplikasikan untuk memprediksi harga penutup emas hari berikutnya sebagai variabel dependen. Namun, terdapat tiga skenario yang akan dianalisis dengan variabel independen yang berbeda untuk setiap datasetnya. Dataset pertama terdiri dari harga penutup emas 5 hari, 10 hari, 15 hari, dan 20 hari sebelumnya. Kemudian harga penutup, pembuka, tertinggi, dan terendah dari emas hari sebelumnya digunakan pada dataset kedua. Dan dataset terakhir memanfaatkan harga penutup, pembuka, tertinggi, dan terendah dari minyak, nilai tukar, dan saham hari sebelumnya. Setiap model yang diperoleh akan di evaluasi dengan membandingkan nilai dari MAE, MAPE, dan R-squared. Hasil penelitian ini menunjukkan bahwa model yang memiliki akurasi paling baik adalah metode Support Vector Regression dan variabel independen yang paling baik dalam menjelaskan variabel dependen adalah harga penutup emas beberapa hari sebelumnya. / Gold is one of the precious metal which is actively traded by investors and portfolio manager. As a safe haven asset that provides security against market price declines, gold can safeguard its value, contribute to portfolio stabilization, and serve as a source of liquidity in unstable conditions. However, gold’s value fluctuates and undergoes continuous changes over time. Therefore, predicting gold prices is crucial for increasing profits and reducing losses. Predictions can be made using RStudio software and employing data mining concepts in the business and financial fields. The research utilizes Support Vector Regression and Random Forest Regression methods. The data employed consists of daily data on gold (Gold Spot US Dollar), oil (Bloomberg WTI Crude Oil), exchange rates (US Dollar Index), and stocks (S&P 500), including opening, closing, highest, and lowest prices from 2020 to 2022. The study aims to predict the next day’s closing gold price as the dependent variable. However, three scenarios with different independent variables for each dataset are analyzed. The first dataset comprises closing prices of gold from 5, 10, 15, and 20 days prior. The second dataset utilizes the closing, opening, highest, and lowest prices of gold from the previous day. The third dataset incorporates the closing, opening, highest, and lowest prices of oil, exchange rates, and stocks from the previous day. Each obtained model is evaluated by comparing the values of MAE, MAPE, and R-squared. The research results indicate that the Support Vector Regression method achieves the highest accuracy, and the independent variable that best explains the dependent variable is the closing price of gold from several days prior.

Item Type: Thesis (Bachelor)
Creators:
CreatorsNIMEmail
Satyawan, ElisaNIM01112190014elisa.satyawan@gmail.com
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorWidjaja, PetrusNIDN0314095901petrus.widjaja@uph.edu
Thesis advisorCahyadi, LinaNIDN0328077701lina.cahyadi@uph.edu
Uncontrolled Keywords: support vector regression ; random forest regression ; harga emas ; prediksi
Subjects: Q Science > QA Mathematics
Divisions: University Subject > Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Mathematics
Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Mathematics
Depositing User: Elisa Elisa
Date Deposited: 02 Feb 2024 01:34
Last Modified: 02 Feb 2024 01:34
URI: http://repository.uph.edu/id/eprint/61355

Actions (login required)

View Item View Item