Implementasi Machine Learning untuk Prediksi Imbal Hasil Obligasi Indonesia

Azizah, Leyli Lathifatul (2021) Implementasi Machine Learning untuk Prediksi Imbal Hasil Obligasi Indonesia. Other thesis, Institut Teknologi Sepuluh Nopember.

	Text ilovepdf_merged.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (2MB) | Request a copy
	Text 06111940000076-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (2MB) | Request a copy

Abstract

Imbal hasil obligasi yang diperoleh investor mengalami perubahan dari waktu ke waktu. Perubahan imbal hasil tersebut berpengaruh pada tingkat harga pasar obligasi itu sendiri. Oleh karena itu, investor dan pemerintah harus selalu memperhatikan fluktuasi harga imbal hasil obligasi tersebut. Studi penelitian terkait prediksi imbal hasil obligasi telah menggunakan model machine learning dengan tingkat akurasi yang baik. Akan tetapi, masih belum diterapkan untuk prediksi imbal hasil obligasi Indonesia. Untuk permasalahan prediksi imbal hasil obligasi Indonesia, penelitian ini menggunakan model machine learning berupa Support Vector Regression (SVR), Multilayer Perceptron (MLP), dan Transformer. Tujuan dari penelitian ini adalah untuk mendapatkan nilai prediksi yang paling akurat dari ketiga model machine learning tersebut. Dataset yang digunakan untuk uji nilai akurasi diperoleh dari data imbal hasil obligasi Indonesia untuk tenor 10 tahun dalam kurun waktu 2017-2022. Untuk prediksi short term seminggu kedepan dan long term selama 5 tahun kedepan, didapatkan hasil terbaik menggunakan metode MLP dengan tingkat akurasi short term untuk MAE sebesar 0.07796, MAPE sebesar 0.01171, dan RMSE sebesar 0.10947 dan tingkat akurasi long term untuk MAE sebesar 0.55166, MAPE sebesar 0.07716, dan RMSE sebesar 0.66122.
=================================================================================================================================
Bond yields obtained by investors have changed from time to time. Changes in yields affect the market price level of the bonds themselves. Therefore, investors and the government must always pay attention to fluctuations in the price of these bond yields. Research studies related to the prediction of bond yields have used machine learning models with a good level of accuracy. However, it has not yet been applied to predictions of Indonesian bond yields. For the problem of predicting Indonesian bond yields, this study uses machine learning models in the form of Support Vector Regression (SVR), Multilayer Perceptron (MLP), and Transformers. The purpose of this research is to get the most accurate predictive value of the three machine learning models. The dataset used to test the accuracy value was obtained from data on Indonesian bond yields for a tenor of 10 years in the period 2017-2022. For short term predictions for the next week and long term for the next 5 years, the best results are obtained using the MLP method with an accuracy level for short term for MAE 0.07796, MAPE 0.01171, and RMSE 0.10947. In addition, an accuracy level of long term for MAE 0.55166, MAPE 0.07716, and RMSE 0.66122.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	imbal hasil obligasi, peramalan, multilayer perceptron, support vector regression, transformer Keywords: bond yield, forecasting, multilayer perceptron, support vector regression, transformer
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. R Medicine > R Medicine (General) > R858 Deep Learning T Technology > T Technology (General) > T174 Technological forecasting T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Chemistry > 47201-(S1) Undergraduate Thesis
Depositing User:	Leyli Lathifatul Azizah
Date Deposited:	23 Sep 2024 03:44
Last Modified:	23 Sep 2024 03:44
URI:	http://repository.its.ac.id/id/eprint/103356

Actions (login required)

View Item