Safira, Alya Mirza (2025) PREDIKSI HARGA SAHAM DI INDONESIA MENGGUNAKAN MODEL EXTREME GRADIENT BOOSTING - ADAPTIVE PARTICLE SWARM OPTIMIZATION. Undergraduate thesis, Universitas Pembangunan Nasional Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
21083010039_COVER.pdf Download (878kB) |
|
Text (Bab 1)
21083010039_BAB 1.pdf Download (323kB) |
|
Text (Bab 2)
21083010039_BAB 2.pdf Restricted to Registered users only until 18 June 2027. Download (661kB) |
|
Text (Bab 3)
21083010039_BAB 3.pdf Restricted to Registered users only until 18 June 2027. Download (503kB) |
|
Text (Bab 4)
21083010039_BAB 4.pdf Restricted to Registered users only until 18 June 2027. Download (1MB) |
|
Text (Bab 5)
21083010039_BAB 5.pdf Download (251kB) |
|
Text (Daftar Pustaka)
21083010039_DAFTAR PUSTAKA.pdf Download (201kB) |
|
Text (Lampiran)
21083010039_LAMPIRAN.pdf Restricted to Registered users only until 18 June 2027. Download (316kB) |
Abstract
Fluctuating stock prices are a major challenge in investment decision-making, especially for stocks with high volatility. This study aims to build a stock price prediction model using the Extreme Gradient Boosting (XGBoost) algorithm, which is optimised using the Adaptive Particle Swarm Optimization (APSO) method. The study focuses on two stocks with high volatility, namely PT Jaya Agra Wattie Tbk. (JAWA.JK) and PT Charnic Capital Tbk. (NICK.JK), using historical closing price data from August 2019 to July 2024. The research process includes data collection, preprocessing, modelling using the Extreme Gradient Boosting (XGBoost) method, optimization using the Adaptive Particle Swarm Optimization (APSO) method, and model performance evaluation using the Mean Absolute Percentage Error (MAPE) metric. The XGBoost-APSO model provided the best prediction performance for JAWA.JK and NICK.JK stocks, with MAPE values of 0.03 (training) and 0.04 (testing) for JAWA.JK, and 0.02 (training) and 0.04 (testing) for NICK.JK. The MAPE The XGBoost-PSO model for JAWA.JK stock is 2.52 (training) and 4.47 (testing), while NICK.JK has a MAPE value of 1.84 (training) and 3.59 (testing). Meanwhile, the XGBoost model without optimization showed the lowest performance among the three models with a MAPE of 3.80 (training) and 4.69 (testing) for JAWA.JK, and 3.88 (training) and 3.99 (testing) for NICK.JK. This model also successfully Predicted the movement of closing prices in the next week realistically according to historical volatility characteristics. This study proves that the combination of XGBoost with APSO optimization can significantly improve the accuracy and stability of the model in stock price prediction and can be used as a predictive tool in investment decision-making. Keywords : Adaptive Particle Swarm Optimization (APSO), Extreme Gradient Boosting (XGBoost), Prediction, Stock, Volatility.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76.6 Computer Programming |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Data Science | ||||||||||||
| Depositing User: | Alya Mirza Safira | ||||||||||||
| Date Deposited: | 19 Jun 2025 01:40 | ||||||||||||
| Last Modified: | 19 Jun 2025 01:40 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/38483 |
Actions (login required)
 |
View Item |