OPTIMASI PORTOFOLIO SAHAM IDX30 MENGGUNAKAN ADAPTIVE GENETIC ALGORITHM DENGAN EVALUASI WALK-FORWARD OPTIMIZATION

Miftahullah Surya Nugraha, NIM.: 22106050038 (2026) OPTIMASI PORTOFOLIO SAHAM IDX30 MENGGUNAKAN ADAPTIVE GENETIC ALGORITHM DENGAN EVALUASI WALK-FORWARD OPTIMIZATION. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (OPTIMASI PORTOFOLIO SAHAM IDX30 MENGGUNAKAN ADAPTIVE GENETIC ALGORITHM DENGAN EVALUASI WALK-FORWARD OPTIMIZATION) 22106050038_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (5MB) | Preview
	Text (OPTIMASI PORTOFOLIO SAHAM IDX30 MENGGUNAKAN ADAPTIVE GENETIC ALGORITHM DENGAN EVALUASI WALK-FORWARD OPTIMIZATION) 22106050038_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (15MB) | Request a copy

Abstract

Portfolio optimization is a complex problem that involves balancing return, risk, and diversification, while the conventional Standard Genetic Algorithm (SGA) is still prone to premature convergence due to its fixed crossover and mutation probabilities. This study aims to optimize an IDX30 stock portfolio using the Adaptive Genetic Algorithm (AGA) evaluated through the Walk-Forward Optimization (WFO) approach. The dataset consists of daily closing prices of IDX30 constituent stocks from January 2023 to January 2026, including only stocks that consistently remained in the IDX30 index throughout the observation period. The research methodology comprises data preprocessing, expected return and covariance matrix estimation, implementation of SGA and AGA, hyperparameter tuning using Grid Search, and out-of-sample evaluation using WFO with SGA, the Equal Weight strategy, and the IDX30 index as benchmarks. The experimental results demonstrate that AGA achieved the best portfolio performance, with a Total Return of 46.49%, an Annual Return of 50.59%, annual volatility of 21.21%, and a Sharpe Ratio of 2.10. Furthermore, AGA produced the lowest Maximum Drawdown of −15.74% and exhibited better solution quality and optimization stability than SGA. These findings indicate that the Adaptive Genetic Algorithm is capable of generating a more optimal, diversified, and adaptive portfolio than the benchmark methods under dynamic market conditions.

Share this knowledge with your friends :

Actions (login required)

View Item