Optimisasi Portofolio Saham Jakarta Islamic Index Berbasis Mean Value-At-Risk Menggunakan Particle Swarm Optimization dan Fireworks Algorithm

Ramadhan, Muhammad Ilham (2026) Optimisasi Portofolio Saham Jakarta Islamic Index Berbasis Mean Value-At-Risk Menggunakan Particle Swarm Optimization dan Fireworks Algorithm. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003221185-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (7MB) | Request a copy

Abstract

Pertumbuhan signifikan jumlah investor saham syariah di Indonesia mendorong kebutuhan akan strategi pengelolaan portofolio yang akurat terhadap karakteristik pasar modal syariah. Data return saham Jakarta Islamic Index (JII) menunjukkan fenomena distribusi non-normal dengan fat-tail dan volatility clustering yang menyebabkan model Mean-Variance klasik tidak mampu menangkap risiko secara memadai sehingga diperlukan pendekatan Mean-Value at Risk (Mean-VaR) sebagai alternatif pengukuran yang lebih representatif. Namun, sifat non-konveks dari fungsi objektif Mean-VaR menuntut penggunaan algoritma metaheuristik sebagai solusi optimasi yang efektif. Penelitian ini menggunakan data sekunder berupa harga penutupan harian 18 saham JII yang konsisten terdaftar selama periode Desember 2022 hingga November 2025. Karakteristik distribusi return diidentifikasi melalui uji goodness-of-fit Anderson-Darling yang menunjukkan bahwa 13 saham mengikuti distribusi Burr (4P), 4 saham mengikuti distribusi Log-Logistic (3P), sedangkan 1 saham tidak memenuhi kedua distribusi tersebut. Berdasarkan estimasi expected return teoritis, dipilih lima saham dengan expected return positif terbesar, yaitu BRMS, BRIS, ANTM, EXCL, dan UNTR sebagai aset penyusun portofolio. Optimasi portofolio dilakukan menggunakan Particle Swarm Optimization (PSO) dan Fireworks Algorithm (FWA) pada model Mean-VaR dengan tiga tingkat preferensi risiko, yaitu konservatif, moderat, dan agresif. Perbandingan performa kedua algoritma dilakukan berdasarkan nilai fitness Mean-VaR. Hasil penelitian menunjukkan bahwa FWA menghasilkan nilai fitness yang lebih tinggi dibandingkan PSO dengan peningkatan hingga 14% pada seluruh preferensi risiko. Sementara itu, PSO memiliki keunggulan dari sisi konsistensi hasil dengan nilai standar deviasi yang mendekati nol serta efisiensi waktu komputasi dengan rata-rata sekitar 5 detik. Seluruh portofolio menghasilkan nilai Reward-to-Value at Risk (RVaR) yang positif, yang menunjukkan bahwa portofolio mampu memberikan excess return di atas aset bebas risiko setelah memperhitungkan risiko ekstrem. Dengan demikian, Fireworks Algorithm (FWA) merupakan metode yang lebih efektif dalam optimasi portofolio Mean-VaR pada saham Jakarta Islamic Index, sedangkan berdasarkan ukuran RVaR, portofolio konservatif memiliki efisiensi imbal hasil terhadap risiko ekstrem yang paling tinggi dengan nilai sebesar 3,7379.
==================================================================================================================================
The significant growth in the number of Islamic stock investors in Indonesia has increased the need for portfolio management strategies that reflect the characteristics of the Islamic capital market. The return data of stocks listed in the Jakarta Islamic Index (JII) exhibit non-normal distributions characterized by fat tails and volatility clustering, making the classical Mean-Variance model inadequate for measuring risk. Therefore, the Mean-Value at Risk (Mean-VaR) model is adopted as a more representative approach. However, the non-convex nature of the Mean-VaR objective function requires metaheuristic algorithms to obtain optimal solutions. This study uses the daily closing prices of 18 JII stocks consistently listed from December 2022 to November 2025. Return distributions were identified using the Anderson-Darling goodness-of-fit test, indicating that 13 stocks followed the Burr (4P) distribution, four followed the Log-Logistic (3P) distribution, while one stock did not fit either distribution. Based on the estimated theoretical expected returns, the five stocks with the highest positive expected returns, namely BRMS, BRIS, ANTM, EXCL, and UNTR, were selected as portfolio assets. Portfolio optimization was performed using Particle Swarm Optimization (PSO) and the Fireworks Algorithm (FWA) under the Mean-VaR model for three investor risk preferences: conservative, moderate, and aggressive. Algorithm performance was evaluated using the Mean-VaR fitness value. The results show that FWA outperformed PSO across all risk preferences, achieving fitness improvements of up to 14%. Meanwhile, PSO demonstrated better solution consistency with near-zero standard deviations and higher computational efficiency, requiring an average computation time of approximately 5 seconds. The optimal portfolios obtained from FWA were subsequently evaluated using the Reward-to-Value-at-Risk (RVaR) measure. All portfolios produced positive RVaR values, indicating their ability to generate excess returns over the risk-free asset after accounting for extreme risk. Among the three investor preferences, the conservative portfolio achieved the highest RVaR value of 3.7379, indicating the highest risk-return efficiency. Therefore, FWA is identified as the most effective metaheuristic algorithm for solving the Mean-VaR portfolio optimization problem for JII stocks.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Fireworks Algorithm, Jakarta Islamic Index, Mean-Value at Risk, Optimasi Portofolio, Particle Swarm Optimization, Fireworks Algorithm, Jakarta Islamic Index, Mean-Value at Risk, Particle Swarm Optimization, Portfolio Optimization
Subjects:	H Social Sciences > HD Industries. Land use. Labor > HD61 Risk Management H Social Sciences > HG Finance > HG4012 Mathematical models H Social Sciences > HG Finance > HG4529 Investment analysis H Social Sciences > HG Finance > HG4529.5 Portfolio management T Technology > T Technology (General) > T57.84 Heuristic algorithms.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Ilham Ramadhan
Date Deposited:	31 Jul 2026 07:55
Last Modified:	31 Jul 2026 07:55
URI:	http://repository.its.ac.id/id/eprint/140181

Actions (login required)

View Item