Peramalan Harga Saham PT Aneka Tambang Tbk Menggunakan Support Vector Regression Dengan Fruit Fly Optimization Algorithm

Prihastari, Rindy Retno (2026) Peramalan Harga Saham PT Aneka Tambang Tbk Menggunakan Support Vector Regression Dengan Fruit Fly Optimization Algorithm. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2043221021-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Fluktuasi harga saham yang tinggi dan sulit diprediksi menjadi tantangan utama dalam investasi pasar modal. Kondisi ini berdampak signifikan terhadap pengambilan keputusan investor, sehingga diperlukan analisis yang tepat untuk meminimalkan risiko kerugian. PT Aneka Tambang Tbk sebagai salah satu emiten BUMN di sektor energi dan bahan tambang mengalami pergerakan harga saham yang dinamis akibat berbagai faktor, seperti harga emas, nilai tukar, serta volume perdagangan dimana menunjukkan prospek yang menjanjikan meski tetap mengalami volatilitas akibat dinamika global. Oleh karena itu, diperlukan metode peramalan yang mampu menghasilkan prediksi harga saham yang akurat untuk mendukung pengambilan keputusan investasi. Penelitian ini menggunakan metode Support Vector Regression (SVR) yang di optimasi dengan Fruit Fly Optimization Algorithm (FOA) untuk meramalkan harga high saham PT Aneka Tambang Tbk. Model peramalan terbaik untuk harga high menggunakan kernel RBF dengan parameter C (constant) sebesar 28,67749, ε (epsilon) sebesar 0,02948, dan γ (gamma) sebesar 0,08596 dengan nilai MAPE sebesar 2,05983% yang menunjukkan tingkat akurasi peramalan yang sangat baik. Hasil Permutation Feature Importance menunjukkan bahwa variabel harga high saham satu hari sebelumnya merupakan variabel yang paling berpengaruh dalam proses prediksi. Selain itu, hasil peramalan selama 22 periode menunjukkan bahwa harga high saham PT Aneka Tambang Tbk masih berfluktuasi dengan kecenderungan meningkat. Data aktual juga berada dalam rentang confidence interval 95%, yang menunjukkan bahwa model mampu merepresentasikan pola pergerakan harga saham dengan baik.
================================================================================================================================
High and unpredictable stock price fluctuations are a major challenge in capital market investment. This condition significantly impacts investor decision-making, so proper analysis is needed to minimize the risk of loss. PT Aneka Tambang Tbk, as one of the state-owned issuers in the energy and mining sector, experiences dynamic stock price movements due to various factors, such as gold prices, exchange rates, and trading volume, which show promising prospects despite still experiencing volatility due to global dynamics. Therefore, a forecasting method is needed that can produce accurate stock price predictions to support investment decision-making. This study uses the Support Vector Regression (SVR) method optimized with the Fruit Fly Optimization Algorithm (FOA) to predict the high price of PT Aneka Tambang Tbk shares. The best forecasting model for the high price uses the RBF kernel with parameters C (constant) of 28.67749, ε (epsilon) of 0.02948, and γ (gamma) of 0.08596 with a MAPE value of 2.05983%, indicating a very good level of forecasting accuracy. The Permutation Feature Importance results indicate that the previous day's high stock price is the most influential variable in the prediction process. Furthermore, the forecasting results for 22 periods indicate that the high stock price of PT Aneka Tambang Tbk continues to fluctuate with an upward trend. The actual data also falls within the 95% confidence interval, indicating that the model adequately represents stock price movement patterns.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Fruit Fly Optimization Algorithm, Harga Saham, Peramalan, Support Vector Regression. Fruit Fly Optimization Algorithm, Forecasting, Stock Price, Support Vector Regression.
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > Q Science (General) > Q337.3 Swarm intelligence Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression Q Science > QA Mathematics > QA353.K47 Kernel functions (analysis) Q Science > QA Mathematics > QA9.58 Algorithms
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Rindy Retno Prihastari
Date Deposited:	16 Jul 2026 08:22
Last Modified:	16 Jul 2026 08:22
URI:	http://repository.its.ac.id/id/eprint/135240

Actions (login required)

View Item