Vannessa, Dhevinta (2024) Perbandingan kinerja model machine learning dalam memprediksi harga minyak mentah Brent = Comparison of machine learning model perfrormance in forecasting Brent crude oil prices. Bachelor thesis, Universitas Pelita Harapan.
![Title [thumbnail of Title]](http://repository.uph.edu/style/images/fileicons/text.png)
Title.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (344kB)
![Abstract [thumbnail of Abstract]](http://repository.uph.edu/style/images/fileicons/text.png)
Abstract.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (1MB)
![ToC [thumbnail of ToC]](http://repository.uph.edu/style/images/fileicons/text.png)
ToC.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (823kB)
![Chapter1 [thumbnail of Chapter1]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter1.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (618kB)
![Chapter2 [thumbnail of Chapter2]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (946kB)
![Chapter3 [thumbnail of Chapter3]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (814kB)
![Chapter4 [thumbnail of Chapter4]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter4.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (8MB)
![Chapter5 [thumbnail of Chapter5]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter5.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (564kB)
![Bibliography [thumbnail of Bibliography]](http://repository.uph.edu/style/images/fileicons/text.png)
Bibliography.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (579kB)
![Appendices [thumbnail of Appendices]](http://repository.uph.edu/style/images/fileicons/text.png)
Appendices.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (46MB)
Abstract
Minyak mentah merupakan salah satu komoditas yang sangat penting di dunia karena pergerakan harga minyak mentah dapat mempengaruhi perekonomian global. Prediksi harga minyak mentah yang akurat dapat membantu pemerintah, perusahaan publik dan swasta, serta para investor dalam mengambil keputusan yang tepat. Tujuan dilakukan penelitian ini adalah untuk membandingkan kinerja model SVR dan RF dalam memprediksi harga penutupan minyak mentah Brent. Data yang digunakan adalah data harga spot minyak mentah Brent yang diambil secara harian dengan rentang waktu 1 Juli 2018 sampai 20 Januari 2024. Model yang digunakan untuk memprediksi harga penutupan minyak mentah adalah model SVR dan RF. Evaluasi model yang digunakan untuk menentukan model yang paling baik adalah MAE, MAPE, dan R-squared. Model SVR dan RF menunjukkan kinerja paling baik ketika diterapkan pada dataset selama pandemi COVID-19 dengan model SVR menggunakan sembilan variabel independen dan model RF menggunakan tujuh variabel independen. Hasil prediksi dengan menggunakan model SVR memiliki tingkat kesalahan prediksi yang rendah dengan nilai MAE sebesar 1,43373 atau sama dengan nilai MAPE sebesar 2,75295%, serta nilai R-squared sebesar 99,168%. Sedangkan, hasil prediksi model RF juga memiliki tingkat kesalahan prediksi yang rendah dengan nilai MAE sebesar 1,33413 atau sama dengan nilai MAPE sebesar 1,96693%, serta nilai R-squared sebesar 99,339%. Model yang paling baik dari dua model yang dibangun untuk memprediksi harga penutupan minyak mentah Brent adalah model RF. / Crude oil is one of the most important commodities in the world because the movement of crude oil prices can affect the global economy. Accurate prediction of crude oil prices can assist governments, public and private companies, as well as investors in making informed decisions. The purpose of this research is to compare the performance of SVR and RF models in predicting Brent crude oil closing prices. The data used is the daily spot price data of Brent crude oil taken from July 1, 2018, to January 20, 2024. The models used to predict Brent crude oil closing prices are SVR and RF models. The evaluation of the models used to determine the best model includes MAE, MAPE, and R-squared. SVR and RF models show the best performance when applied to the dataset during the COVID-19 pandemic, with the SVR model using nine independent variables and the RF model using seven independent variables. The prediction results using the SVR model have a low prediction error rate with an MAE value of 1.43373 or an equivalent MAPE value of 2.75295%, and an R-squared value of 99.168%. Meanwhile, the prediction results of the RF model also have a low prediction error rate with an MAE value of 1.33413 or an equivalent MAPE value of 1.96693%, and an R-squared value of 99.339%. The best model among the two models built to predict Brent crude oil closing prices is the RF model.
Item Type: | Thesis (Bachelor) |
---|---|
Creators: | Creators NIM Email ORCID Vannessa, Dhevinta NIM01112200021 dhevintavannessa@gmail.com UNSPECIFIED |
Contributors: | Contribution Contributors NIDN/NIDK Email Thesis advisor Cahyadi, Lina NIDN0328077701 lina.cahyadi@uph.edu Thesis advisor Widjaja, Petrus NIDN0314095901 petrus.widjaja@uph.edu |
Uncontrolled Keywords: | random forest; support vector regression; machine learning; harga minyak mentah; kinerja model; random forest; support vector regression; machine learning; crude oil prices; model performance. |
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: | Dhevinta Vannessa |
Date Deposited: | 19 Jul 2024 03:38 |
Last Modified: | 19 Jul 2024 03:38 |
URI: | http://repository.uph.edu/id/eprint/64000 |