Penerapan Machine Learning Algoritma Random Forest Dalam Menganalisis Dampak Rasio Perbankan Terhadap Pergerakan Saham

Diantini, Ni Luh Putu Ayu

JURNAL TEKNIK INFORMATIKA DAN SISTEM INFORMASI
Vol 12 No 2 (2025): JATISI (Jurnal Teknik Informatika dan Sistem Informasi)

This research examines the influence of bank-specific financial ratios on stock price movements of Indonesia’s Big Four banks (BBCA, BBRI, BMRI, and BBNI) using a machine learning approach with the Random Forest algorithm. The research utilizes the Cross-Industry Standard Process for Data Mining (CRISP-DM) framework using quarterly data from Q2 2019 to Q2 2024, which consists of 6 key financial ratios, namely Return on Equity (ROE), Return on Assets (ROA), Net Interest Margin (NIM), Non-Performing Loans (NPL), Loan to Deposit Ratio (LDR), and Capital Adequacy Ratio (CAR), and also the stock price. Results indicate that ROE, CAR, and NPL significantly affect stock prices, with ROE being the most impactful predictor and NIM being the least. The Random Forest model achieved high predictive accuracy, validated by evaluation metrics such as MAPE and R². A practical interface was also developed using Streamlit to facilitate analysis and decision-making. This research highlights the potential of machine learning to enhance financial analysis and provides a foundation for further exploration with expanded datasets and alternative models.