PREDIKSI JAKARTA ISLAMIC INDEX DENGAN METODE EXTREME GRADIENT BOOSTING YANG DIOPTIMALKAN MENGGUNAKAN BAYESIAN OPTIMIZATION

Aryani, Defi (2026) PREDIKSI JAKARTA ISLAMIC INDEX DENGAN METODE EXTREME GRADIENT BOOSTING YANG DIOPTIMALKAN MENGGUNAKAN BAYESIAN OPTIMIZATION. Bachelor thesis, Institut Teknologi Kalimantan.

	Text 17221024_cover.pdf Restricted to Repository staff only until 3 October 2028. Download (594kB) | Request a copy
	Text 17221024_statement_of_authenticity.pdf Restricted to Repository staff only until 3 October 2028. Download (277kB) | Request a copy
	Text 17221024_publishing_agreement.pdf Restricted to Repository staff only until 3 October 2028. Download (271kB) | Request a copy
	Text 17221024_approval_sheet.pdf Restricted to Repository staff only until 3 October 2028. Download (262kB) | Request a copy
	Text 17221024_preface.pdf Restricted to Repository staff only until 3 October 2028. Download (315kB) | Request a copy
	Text 17221024_abstract_id.pdf Restricted to Repository staff only until 3 October 2028. Download (316kB) | Request a copy
	Text 17221024_abstract_en.pdf Restricted to Repository staff only until 3 October 2028. Download (276kB) | Request a copy
	Text 17221024_table_of_content.pdf Restricted to Repository staff only until 3 October 2028. Download (370kB) | Request a copy
	Text 17221024_illustrations.pdf Restricted to Repository staff only until 3 October 2028. Download (218kB) | Request a copy
	Text 17221024_tables.pdf Restricted to Repository staff only until 3 October 2028. Download (273kB) | Request a copy
	Text 17221024_notations.pdf Restricted to Repository staff only until 3 October 2028. Download (457kB) | Request a copy
	Text 17221024_chapter_1.pdf Restricted to Repository staff only until 3 October 2028. Download (421kB) | Request a copy
	Text 17221024_chapter_2.pdf Restricted to Repository staff only until 3 October 2028. Download (609kB) | Request a copy
	Text 17221024_chapter_3.pdf Restricted to Repository staff only until 3 October 2028. Download (450kB) | Request a copy
	Text 17221024_chapter_4.pdf Restricted to Repository staff only until 3 October 2028. Download (787kB) | Request a copy
	Text 17221024_conclusions.pdf Restricted to Repository staff only until 3 October 2028. Download (311kB) | Request a copy
	Text 17221024_bibliography.pdf Restricted to Repository staff only until 3 October 2028. Download (282kB) | Request a copy
	Text 17221024_enclosure.pdf Restricted to Repository staff only until 3 October 2028. Download (577kB) | Request a copy
	Text 17221024_paper.pdf Restricted to Repository staff only until 3 October 2028. Download (835kB) | Request a copy
	Text 17221024_presentation.pdf Restricted to Repository staff only until 3 October 2028. Download (1MB) | Request a copy
	Text 17221024_Form. TA-020.pdf Restricted to Repository staff only until 3 October 2028. Download (561kB) | Request a copy

Abstract

Jakarta Islamic Index (JII) merupakan indikator utama saham syariah di Indonesia yang memiliki tingkat volatilitas tinggi dan pola data deret waktu non-linear, sehingga sulit diprediksi menggunakan metode konvensional. Penelitian ini bertujuan untuk memprediksi pergerakan JII menggunakan algoritma Extreme Gradient Boosting (XGBoost) serta meningkatkan kinerjanya melalui optimasi hyperparameter menggunakan Bayesian Optimization. Data yang digunakan berupa harga penutupan harian JII periode Januari 2020 hingga Desember 2025 dengan proporsi pembagian data latih dan data uji sebesar 80:20. Kinerja model dievaluasi menggunakan Mean Absolute Percentage Error (MAPE) dan Root Mean Square Error (RMSE). Hasil penelitian menunjukkan bahwa model XGBoost baseline memperoleh nilai RMSE sebesar 33,18 dan MAPE sebesar 4,69% sedangkan setelah dilakukan optimasi menggunakan Bayesian Optimization nilai RMSE menurun menjadi 30,72 dan MAPE menjadi 4,17%. Hal ini menunjukkan bahwa hasil optimasi hyperparameter mampu meningkatkan akurasi model dalam memprediksi pergerakan Jakarta Islamic Index, meskipun peningkatan yang diperoleh relatif kecil. Pendekatan ini diharapkan dapat menjadi alternatif yang andal dalam mendukung pengambilan keputusan investasi di pasar modal syariah. Secara keseluruhan, kombinasi algoritma XGBoost dan Bayesian Optimization menunjukkan potensi yang baik dalam pemodelan data time series keuangan, khususnya untuk prediksi indeks saham syariah di Indonesia.

Item Type:	Thesis (Bachelor)
Subjects:	Q Science > QA Mathematics
Divisions:	Jurusan Matematika dan Teknologi Informasi > Ilmu Aktuaria
Depositing User:	Defi Aryani
Date Deposited:	14 Jul 2026 01:02
Last Modified:	14 Jul 2026 01:02
URI:	http://repository.itk.ac.id/id/eprint/27049

Actions (login required)

View Item