Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Jurnal Ilmiah KOMPUTASI
I Made Tirta
Unknown Affiliation
Computer Science & IT

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Prediksi Harga Saham PT Bank Rakyat Indonesia Tbk Menggunakan AUTOML H2O I Made Tirta; Abduh Riski; Sholikhah, Nining
Jurnal Ilmiah Komputasi Vol. 23 No. 3 (2024): Jurnal Ilmiah Komputasi : Vol. 23 No 3, September 2024
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32409/jikstik.23.3.3624

Abstract

Bank BRI is a government-owned company with share prices recorded in the Initial Public Offering (IPO) which has the status of a public company. BRI Bank's share price experienced fluctuations caused by some factors. Predicting BRI Bank share prices is important to make it easier for investors to enter make investment decisions. Auto Machine Learning (AutoML) refers to the concept of machine learning and training automatic parameter setting. H2OAutoML can be used to predict stock prices with deliver program code and accelerate the development of accurate algorithms. H2OAutoML provides various algorithms, but the one used in this research is the Generalized Linear Model (GLM), Distributed Random Forest (DRF), Gradient Boosting Machine (GBM), and stacked ensemble. The aim of this research is to find out the optimal algorithm and prediction results produced by H2OAutoML on close stock prices. Algorithm The best basis according to H2OAutoML is GBM with the smallest MAPE value and the largest R Square. However, when this basic algorithm combined with stacking techniques produces better predictions. The basic algorithm used to build stacked ensembles are DRF, XRT, GLM, and GBM. This stacked ensemble is constructed sequentially automatically by H2OAutoML with the GLM metalearning algorithm. Thus, stacked ensembles are capable predicts with fairly good accuracy and can explain data variability.
Co-Authors Abduh Riski Sholikhah, Nining