Pradani, Azizah Jois (2026) Implementasi Artificial Bee Colony Algorithm untuk Optimasi Portofolio Saham Pefindo i-Grade. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 2043221144_Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
2043221144_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy |
Abstract
Investasi saham menawarkan potensi return yang cukup tinggi namun dihadapkan oleh risiko volatilitas harga yang dipengaruhi oleh ketidakpastian ekonomi global. Oleh karena itu, diperlukan pengelolaan portofolio agar investor dapat memperoleh return maksimal dengan risiko yang terkendali. Salah satu metode optimasi portofolio yang sering digunakan dalam manajemen portofolio adalah model mean-variance. Tetapi dalam praktiknya, model ini dihadapkan dengan kendala realistis seperti buy-in threshold, cardinality constraint, dan round lot. Permasalahan optimasi tersebut akhirnya termasuk dalam kategori Non-deterministic Polynomial-time hard (NP-hard), sehingga memerlukan pendekatan yang lebih kompleks seperti algoritma metaheuristic. Oleh karena itu, penelitian ini bertujuan untuk membentuk portofolio saham optimal dengan saham Indeks Pefindo i-Grade menggunakan algoritma Artificial Bee Colony (ABC) yang mampu menangani kendala realistis berupa buy-in threshold, cardinality constraint, dan roundlot constraint. Data yang digunakan adalah harga penutupan harian dari 30 saham yang tergabung dalam Indeks Pefindo i-Grade selama periode Januari hingga Juni 2025. Hasil penelitian menunjukkan bahwa setelah dilakukan eliminasi saham berdasarkan expected return positif, proses optimasi portofolio dilanjutkan pada 13 skenario jumlah saham dalam portofolio menggunakan algoritma Artificial Bee Colony. Pemilihan portofolio optimal dilakukan dengan nilai Sharpe Ratio. Dari seluruh skenario yang diuji, portofolio dengan enam saham (k = 6) menghasilkan kinerja terbaik dengan nilai Sharpe Ratio tertinggi sebesar 0,127. Portofolio optimal tersebut tersusun atas saham ADHI, HRTA, MBMA, PTRO, TOBA, dan TPIA, dengan distribusi bobot yang telah memenuhi seluruh kendala investasi yang ditetapkan. Selain itu, analisis korelasi antar saham menunjukkan bahwa portofolio yang terbentuk memiliki tingkat diversifikasi yang baik. Temuan ini membuktikan bahwa algoritma Artificial Bee Colony efektif dalam menyelesaikan permasalahan optimasi portofolio yang bersifat NP-Hard dan menghasilkan portofolio optimal bagi investor.
================================================================================================================================
Stock investment offers potential for high returns but is faced with price volatility risks influenced by global economic uncertainty. Therefore, portfolio management is essential to enable investors to obtain maximum returns with controlled risk. One of the most frequently used methods in portfolio management is the mean-variance model. However, in practice, this model faces realistic constraints such as buy-in threshold, cardinality constraint, and round lot. Such optimization challenges are classified as Non-deterministic Polynomial-time hard (NP-hard) problems, necessitating advanced computational approaches such as metaheuristic algorithms. Therefore, this study aims to construct an optimal stock portfolio from the PEFINDO i-Grade Index using the Artificial Bee Colony (ABC) algorithm, which can handle realistic constraints, namely buy-in threshold, cardinality constraint, and round lot constraint. The data used consists of daily closing prices of 30 stocks within the PEFINDO i-Grade Index for the period of January to June 2025. The results show that after eliminating stocks based on positive expected returns, the portfolio optimization process continued through 13 stock-count scenarios using the Artificial Bee Colony algorithm. Portfolio selection was adjudicated based on the Sharpe Ratio to evaluate risk-adjusted performance. The empirical results demonstrate that a six-stock configuration (k = 6) yielded the highest performance, achieving the highest Sharpe Ratio of 0.127. This optimal portfolio is composed of ADHI, HRTA, MBMA, PTRO, TOBA, and TPIA, with weight distributions that fulfill all established investment constraints. Furthermore, correlation analysis between stocks indicates that the formed portfolio possesses a good level of diversification. These findings prove that the Artificial Bee Colony algorithm is effective in solving NP-Hard portfolio optimization problems and generating optimal portfolios for investors.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Artificial Bee Colony, Buy-in Threshold, Kardinalitas, Mean-Variance, Roundlot Artificial Bee Colony, Buy-in Threshold, Cardinality, Mean-Variance, Roundlot |
| Subjects: | H Social Sciences > HG Finance H Social Sciences > HG Finance > HG4529 Investment analysis H Social Sciences > HG Finance > HG4529.5 Portfolio management H Social Sciences > HG Finance > HG4910 Investments Q Science Q Science > QA Mathematics Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA402.5 Genetic algorithms. Interior-point methods. Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Faculty of Vocational > 49501-Business Statistics |
| Depositing User: | Azizah Jois Pradani |
| Date Deposited: | 12 Feb 2026 06:15 |
| Last Modified: | 12 Feb 2026 06:15 |
| URI: | http://repository.its.ac.id/id/eprint/132395 |
Actions (login required)
 |
View Item |