Optimisasi Portofolio Saham IDX30 dengan Conditional Value at Risk Menggunakan Kendala Buy-in Threshold, Kardinalitas, dan Round Lot

Rajagukguk, Dionaldo (2026) Optimisasi Portofolio Saham IDX30 dengan Conditional Value at Risk Menggunakan Kendala Buy-in Threshold, Kardinalitas, dan Round Lot. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5006221037-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Penelitian ini dilakukan untuk merancang model optimisasi portofolio saham IDX30 yang mampu meminimalkan risiko kerugian ekstrem dengan menggunakan Conditional Value at Risk (CVaR) sebagai ukuran risiko utama dan mengintegrasikannya dengan kendala praktis berupa buy-in threshold, kardinalitas, dan aturan round lot yang mencerminkan kondisi nyata pasar modal Indonesia. Latar belakang penelitian ini berangkat dari kebutuhan investor untuk memperoleh portofolio yang tidak hanya efisien secara matematis, tetapi juga implementatif dalam praktik perdagangan sehari-hari. Tujuan penelitian ini adalah memformulasikan model Mean-CVaR ke dalam bentuk Mixed Integer Linear Programming (MILP), memperoleh solusi optimal menggunakan metode simpleks dan branch and bound, mengukur kinerja portofolio optimal, serta menganalisis bagaimana perubahan parameter kendala memengaruhi performa portofolio. Metode penelitian melibatkan 17 aset IDX30 dengan 968 skenario return historis, di mana model diselesaikan melalui LP-relaxation dan optimisasi integer secara bertahap. Hasil penelitian menunjukkan bahwa model mampu menghasilkan portofolio optimal dengan nilai CVaR sebesar 0,018672, expected return harian 0,000199, serta sepuluh saham terpilih sesuai kendala kardinalitas. Evaluasi performa juga mencatat sharpe ratio 2,679695, dan cumulative return 0,050036 yang menandakan bahwa portofolio yang terbentuk berada pada kondisi yang stabil dan terukur. Analisis sensitivitas memperlihatkan bahwa penyesuaian nilai α berdampak langsung pada peningkatan CVaR dan perubahan struktur kinerja portofolio, sehingga model ini dapat digunakan sebagai alat bantu pengambilan keputusan yang adaptif terhadap preferensi risiko investor. Kesimpulan penelitian menegaskan bahwa model Mean-CVaR berbasis MILP efektif digunakan untuk optimisasi portofolio dengan mempertimbangkan kendala praktis di pasar modal Indonesia.
======================================================================================================================================
This research aims to design an optimized stock portfolio for the IDX30 index by employing Conditional Value at Risk (CVaR) as the primary risk measure and integrating practical constraints such as the buy-in threshold, cardinality limits, and round-lot trading rules to reflect real market conditions in Indonesia. The study is motivated by the need for investors to construct portfolios that are not only mathematically efficient but also feasible for real-world implementation. The objectives of this research include formulating a Mean–CVaR optimization model within a Mixed Integer Linear Programming (MILP) framework, determining the optimal solution using the simplex method and branch and bound, evaluating the performance of the resulting optimal portfolio, and analyzing how adjustments to constraint parameters affect portfolio outcomes. The model incorporates 17 IDX30 assets and 968 historical return scenarios and is solved through a combination of LP-relaxation and integer optimization procedures. The findings demonstrate that the model successfully produces an optimal portfolio with a CVaR value of 0.018672, a daily expected return of 0.000199, and a selection of ten assets in accordance with the cardinality constraint. Performance evaluation further reveals a sharpe ratio of 2.679695, and a cumulative return of 0.050036, indicating that the constructed portfolio is both stable and well-controlled in terms of risk. Sensitivity analysis shows that changes in the confidence level α directly influence CVaR values and overall performance, suggesting that the model is capable of adapting to varying investor risk preferences. The study concludes that the MILP-based Mean-CVaR model is an effective approach for portfolio optimization under practical constraints within the Indonesian capital market.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Branch and bound, Buy-in threshold, Conditional Value at Risk, Kardinalitas, Round lot, Simpleks. Branch and bound, Buy-in threshold, Conditional Value at Risk, Cardinality, Round lot, Simplex.
Subjects:	Q Science > QA Mathematics > QA275 Theory of errors. Least squares. Including statistical inference. Error analysis (Mathematics) Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) Q Science > QA Mathematics > QA279.5 Bayesian statistical decision theory. Q Science > QA Mathematics > QA401 Mathematical models.
Divisions:	Faculty of Mathematics, Computation, and Data Science > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Dionaldo Rajagukguk
Date Deposited:	12 Jan 2026 05:24
Last Modified:	12 Jan 2026 05:24
URI:	http://repository.its.ac.id/id/eprint/129481

Actions (login required)

View Item