Optimasi Portofolio Berbasis Copula dan Extreme Value Theory untuk Meminimalkan Expected Shortfall dengan Kendala Skor ESG

Meilinda, Alfira Rosa (2026) Optimasi Portofolio Berbasis Copula dan Extreme Value Theory untuk Meminimalkan Expected Shortfall dengan Kendala Skor ESG. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5006221082-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (7MB) | Request a copy

Abstract

Ketidakstabilan pasar keuangan global yang ditandai oleh distribusi return bersifat fat tail dan meningkatnya ketergantungan antar aset pada kondisi ekstrem mendorong perlunya pendekatan pengukuran risiko yang lebih akurat. Penelitian ini bertujuan mengembangkan model optimasi portofolio minimum Expected Shortfall (ES) yang mengintegrasikan Extreme Value Theory (EVT), copula, dan kendala Environmental, Social, and Governance (ESG) pada saham-saham Indeks SRI-KEHATI periode 2023-2025. Dari 25 saham anggota indeks, diperoleh 14 saham yang memenuhi kriteria screening berdasarkan skor ESG. Risiko ekstrem dimodelkan menggunakan metode Peaks Over Threshold dengan distribusi Generalized Pareto Distribution (GPD). Hasil menunjukkan bahwa 13 dari 14 saham memiliki distribusi ekor terbatas, sedangkan TLKM menunjukkan karakteristik heavy tail. Struktur dependensi antar saham dimodelkan menggunakan Gaussian copula dan Student-t copula, dengan Student-t copula terpilih sebagai model terbaik berdasarkan nilai Akaike Information Criterion (AIC). Optimasi portofolio dilakukan menggunakan pendekatan minimum ES berbasis simulasi Monte Carlo. Hasil menunjukkan bahwa batas ESG sebesar 49,36 bersifat non-binding karena portofolio optimal tanpa kendala ESG telah memenuhi batas tersebut. Sebagai pembanding, batas ESG sebesar 60 meningkatkan Expected Shortfall sebesar 9,96% dan meningkatkan return tahunan menjadi 1,22% pada tingkat kepercayaan 95%. Hasil tersebut menunjukkan bahwa peningkatan standar ESG diikuti oleh peningkatan risiko ekstrem dan return portfolio.
=====================================================================================================================================
Global financial market instability, characterized by fat-tailed return distributions and increasing dependence among assets under extreme market conditions, highlights the need for more accurate risk measurement approaches. This study aims to develop a minimum Expected Shortfall (ES) portfolio optimization model that integrates Extreme Value Theory (EVT), copulas, and Environmental, Social, and Governance (ESG) constraints for stocks listed in the SRI-KEHATI Index during the 2023-2025 period. Of the 25 constituent stocks, 14 met the ESG screening criteria. Extreme risk was modeled using the Peaks Over Threshold method with the Generalized Pareto Distribution (GPD). The results indicate that 13 of the 14 stocks exhibit bounded-tail distributions, while TLKM demonstrates heavy-tail characteristics. The dependence structure among stocks was modeled using Gaussian and Student-t copulas, with the Student-t copula selected as the best-fitting model based on the Akaike Information Criterion (AIC). Portfolio optimization was performed using a Monte Carlo simulation-based minimum ES approach. The findings show that the ESG threshold of 49.36 is non-binding, as the unconstrained optimal portfolio already satisfies this requirement. For comparison, an ESG threshold of 60 increases Expected Shortfall by 9.96% while raising the annual return to 1.22% at the 95% confidence level. These results suggest a trade-off between extreme risk and higher ESG targets, where stricter ESG standards are associated with both increased portfolio returns and greater exposure to extreme risk.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Copula, ESG, Expected Shortfall, Extreme Value Theory, Optimasi Portofolio, SRI-KEHATI, Copula, ESG, Expected Shortfall, Extreme Value Theory, PortfolioOptimization, SRI-KEHATI
Subjects:	H Social Sciences > HA Statistics > HA31.7 Estimation H Social Sciences > HG Finance > HG4529 Investment analysis H Social Sciences > HG Finance > HG4529.5 Portfolio management H Social Sciences > HG Finance > HG4915 Stocks--Prices
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Alfira Rosa Meilinda
Date Deposited:	17 Jul 2026 06:22
Last Modified:	17 Jul 2026 06:23
URI:	http://repository.its.ac.id/id/eprint/135285

Actions (login required)

View Item