Open access Jurnal SINTA 3
PERFORMANCE EVALUATION OF THE INDF.JK STOCK PRICE MOVEMENT PREDICTION MODEL USING RANDOM FOREST METHOD WITH GRID SEARCH CROSS VALIDATION OPTIMIZATION
BAREKENG: Jurnal Ilmu Matematika dan Terapan · Vol. 19 No. 3 (2025) · pp. 2155–2168
- ISSN
- 2615-3017, 1978-7227
- Published
- 2025-07-01
- Publisher
- PATTIMURA UNIVERSITY
- Language
- eng
Abstract
Investment in financial instruments in Indonesia has shown significant growth over time, with stocks often being the first choice for investors to invest money. Unfortunately, deciding to buy and sell stocks is not easy. When determining the right time to buy or sell stocks, volatile stock price movements and losses caused by wrong decisions are investors' problems. Thus, it is essential to analyze stock price movement predictions. This study aims to evaluate the prediction model's performance for PT Indofood Sukses Makmur Tbk (INDF.JK) stock price movement in the next 30 days to reduce the risk of possible losses and help the decision-making process. We used the Random Forest method and Grid Search Cross Validation (CV) optimization to form the model. The data used is the closing price of INDF.JK stock for the period January 2, 2014, to December 29, 2023, from Yahoo Finance, which is processed into eight types of stock technical indicators, namely SMA_5, SMA_10, SMA_15, SMA_30, EMA_9, MACD, MACD_Signal, and RSI. The research pipeline includes descriptive statistics, preprocessing, feature and target variables determination, data split, model formation without and with optimization, testing accompanied by performance evaluation, and comparison of the formed model. The results show that the prediction model of INDF. JK's stock price movement in the next 30 days has excellent performance, proven accurate by 90.8% with the application of Random Forest and Grid Search CV. The Random Forest prediction model with Grid Search CV optimization has better performance indicators than the Random Forest model without Grid Search CV optimization, which is shown by the increase of all indicator values. The relative Strength Index is the variable with the best performance for the prediction model. It can be used as the primary consideration for investors when deciding on the buying and selling process of INDF.JK stock in the next 30 days.
Collaboration network of the authors
32 authors · 83 co-authorship links
Hover or tap a name to highlight its co-authors. Double-click to open the author's page.
- Universitas Mercu Buana Yogyakarta
- Affiliation unknown
- Different organisations
- Same organisation
- Affiliation unknown
- Thicker line = more joint articles · bigger circle = more articles
Show as table
| Co-author | Organisation | Joint articles |
|---|---|---|
| Neva Satyahadewi | Affiliation unknown | 11 |
| Wiji Umiati | Affiliation unknown | 3 |
| Hendri Agustono | Affiliation unknown | 3 |
| Rifki Pebriyandi | Affiliation unknown | 3 |
| Risky Gunawan | Affiliation unknown | 3 |
| Cinta Priscillia Maharani | Affiliation unknown | 3 |
| Yundari Yundari | Affiliation unknown | 2 |
| Hendra Perdana | Affiliation unknown | 2 |
| Ray Tamtama | Affiliation unknown | 2 |
| Muhammad Fikri | Universitas Mercu Buana Yogyakarta | 2 |
| Evy Sulitianingsih | Affiliation unknown | 2 |
| Setyo Wira Rizki | Affiliation unknown | 1 |
| Geby Marola | Affiliation unknown | 1 |
| Rita Apriliyanti | Affiliation unknown | 1 |
| Tiffany Ardhitha | Affiliation unknown | 1 |
| Indah Oktaviani | Affiliation unknown | 1 |
| Maria Welita Dwinanda | Affiliation unknown | 1 |
| Dinda Tiara | Affiliation unknown | 1 |
| Lisa Lestari | Affiliation unknown | 1 |
| Aditya Nugraha Pratama | Affiliation unknown | 1 |
| Rifkah Alfiyyah Sary | Affiliation unknown | 1 |
| Dadan Kusnandar | Affiliation unknown | 1 |
| Naomi Nessyana Debataraja | Affiliation unknown | 1 |
| Cristy Framedia Cahyani | Affiliation unknown | 1 |
| Ainun Zakiah | Affiliation unknown | 1 |
| Noerul Hanin | Affiliation unknown | 1 |
| David Jordy Dhandio | Affiliation unknown | 1 |
| Tiara Herlinda Sari | Affiliation unknown | 1 |
Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.