Optimasi Portofolio Saham dengan Batasan Kardinalitas Menggunakan Komodo Mlipir Algorithm: Studi Kasus IDXESGL

Putra, Ramandika Aditya (2025) Optimasi Portofolio Saham dengan Batasan Kardinalitas Menggunakan Komodo Mlipir Algorithm: Studi Kasus IDXESGL. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5002211111-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Penelitian ini membahas optimasi portofolio saham dengan batasan kardinalitas menggunakan Komodo Mlipir Algorithm (KMA) sebagai metode metaheuristik. Studi kasus dilakukan pada saham-saham yang tergabung dalam indeks IDX ESG Leaders (IDXESGL) dengan data harga penutupan bulanan periode Januari 2021 hingga Desember 2023. Model optimasi yang digunakan berbasis Mean-Variance Optimization (MVO) dengan tiga varian model: tanpa batasan kardinalitas, dengan batasan kardinalitas eksplisit, dan dengan batasan kardinalitas sebagai fungsi penalti. Penyelesaian model dilakukan dengan pendekatan eksak (software optimasi) serta pendekatan metaheuristik KMA. Hasil penelitian menunjukkan bahwa pendekatan eksak menghasilkan nilai fungsi objektif minimum yang lebih baik, khususnya pada model tanpa kendala jumlah saham. Namun, KMA terbukti efektif untuk menangani model dengan kendala kombinatorial dan fungsi penalti, serta mampu menghasilkan solusi portofolio yang valid sesuai batasan kardinalitas.
==================================================================================================================================
This study discusses stock portfolio optimization with cardinality constraints using the Komodo Mlipir Algorithm (KMA) as a metaheuristic method. The case study focuses on stocks included in the IDX ESG Leaders (IDXESGL) index, using monthly closing price data from January 2021 to December 2023. The optimization model is based on MeanVariance Optimization (MVO) with three model variants: without cardinality constraints, with explicit cardinality constraints, and with cardinality constraints as a penalty
function. The model was solved using exact approaches (optimization software) and the metaheuristic approach of KMA. The results indicate that the exact methods produced better minimum objective function values, particularly for the model without cardinality constraints. However, KMA proved effective in handling models with combinatorial constraints and penalty functions, and was able to generate valid portfolio solutions that met the cardinality requirements.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Optimasi portofolio, batasan kardinalitas, Komodo Mlipir Algorithm, Mean-Variance Optimization, IDXESGL, Portfolio optimization, cardinality constraint, Komodo Mlipir Algorithm, Mean-Variance Optimization, IDXESGL.
Subjects:	Q Science > QA Mathematics > QA401 Mathematical models.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Ramandika Aditya Putra
Date Deposited:	28 Jan 2026 06:32
Last Modified:	28 Jan 2026 06:32
URI:	http://repository.its.ac.id/id/eprint/130534

Actions (login required)

View Item