Prediksi Harga Saham ANTM Menggunakan Gated Recurrent Unit dengan Optimasi Bayesian

Subairi, Subairi (2025) Prediksi Harga Saham ANTM Menggunakan Gated Recurrent Unit dengan Optimasi Bayesian. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 21081010019-cover.pdf Download (789kB)
	Text (Bab 1) 21081010019-bab1.pdf Download (252kB)
	Text (Bab 2) 21081010019-bab2.pdf Restricted to Repository staff only until 5 December 2027. Download (600kB)
	Text (Bab 3) 21081010019-bab3.pdf Restricted to Repository staff only until 5 December 2027. Download (725kB)
	Text (Bab 4) 21081010019-bab4.pdf Restricted to Repository staff only until 5 December 2027. Download (1MB)
	Text (Bab 5) 21081010019-bab5.pdf Download (239kB)
	Text (Daftar pustaka) 21081010019-daftarpustaka.pdf Download (192kB)
	Text (Lampiran) 21081010019-lampiran.pdf Restricted to Repository staff only Download (247kB)

Abstract

The stock price of ANTM continues to fluctuate due to market conditions, global commodity demand, and broader economic dynamics. Therefore, a prediction method capable of capturing the movement patterns of stock prices accurately is required to minimize potential losses. The historical price data of ANTM is categorized as time series data with non-linear patterns, making the Gated Recurrent Unit (GRU) model effective for predicting such characteristics. However, the performance of GRU is highly influenced by the proper selection of hyperparameters. Thus, this research applies Bayesian Optimization in the process of searching for the optimal hyperparameter configuration, with the search space covering the number of GRU units, learning rate, number of epochs, and batch size. This study utilizes ANTM historical stock price data from the 2015–2025 period, divided into 80% for training and 20% for testing. By using closing price and trading volume as input features, training a 2-layer GRU model, and conducting Bayesian Optimization with 60 iterations and 2-fold cross-validation, the model produces low prediction errors: MAE of 26.807, RMSE of 37.99, and MAPE of 1.69%. Model evaluation was performed by comparing it with a non-optimized GRU model as a related comparative benchmark. The results show that the GRU model optimized with Bayesian Optimization outperforms all comparison models. Therefore, it can be concluded that the GRU model optimized using Bayesian Optimization is effective for predicting ANTM stock prices based on time series data.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Sari, Anggraini Puspita	NIDN0716088605	anggraini.puspita.if@upnjatim.ac.id
Thesis advisor	Mandyartha, Eka Prakarsa	NIDN0725058805	eka_prakarsa.fik@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics > QA76.6 Computer Programming
Q Science > QA Mathematics > QA76.87 Neural computers

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Subairi Subairi

Date Deposited:

05 Dec 2025 08:46

Last Modified:

05 Dec 2025 09:02

URI:

https://repository.upnjatim.ac.id/id/eprint/48018