Hayyu, Annisa' Riskika (2026) Rancang Bangun Aplikasi Portofolio Optimal Saham IDXQ30 Dengan Model Markowitz Dan Pengukuran Risiko Menggunakan Extreme Value Theory. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 2043221141-Undergraduate-Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
2043221141-Undergraduate-Thesis.pdf - Accepted Version Restricted to Repository staff only Download (12MB) | Request a copy |
Abstract
Pasar modal di Indonesia menunjukkan adanya pertumbuhan yang signifikan, dimana investor cenderung mencari saham dengan fundamental kuat seperti pada indeks IDX Quality 30. Meskipun saham-saham dalam indeks ini berkualitas, risiko pasar masih tetap ada sehingga memerlukan adanya diversifikasi portofolio dengan model Markowitz. Meskipun model Markowitz membantu optimasi portofolio, keterbatasannya dalam mengukur risiko ekstrem, seperti fenomena "ekor tebal", memerlukan pendekatan lebih akurat. Oleh karena itu, penelitian ini menggunakan perhitungan Value at Risk dengan metode Extreme Value Theory (EVT), khususnya distribusi Generalized Extreme Value (GEV) dan Generalized Pareto Distribution (GPD), agar hasil estimasi lebih akurat. Hasil penelitian ini diharapkan dapat meningkatkan pemahaman mengenai optimasi portofolio dan manajemen risiko ekstrem, sekaligus memberikan informasi bagi investor dalam pengambilan keputusan investasi. Untuk mendukung pengambilan keputusan investasi yang lebih akurat dan efektif, dirancang aplikasi berbasis website yang menawarkan fungsionalitas untuk mengintegrasikan data saham, optimasi portofolio, dan estimasi risiko yang interaktif. Hasil penelitian menunjukkan bahwa portofolio optimal terbentuk dari komposisi saham ADRO sebesar 24%, BNGA sebesar 23%, NISP sebesar 21%, AMRT sebesar 16%, BRIS sebesar 14%, dan MIKA sebesar 2%. Sementara itu, estimasi risiko terbaik diperoleh menggunakan distribusi Generalized Extreme Value (GEV), dengan nilai VaR pada tingkat kepercayaan 90%, 95%, dan 99% berturut-turut sebesar -2,3769%, -2,5652%, dan 2,8224%. Aplikasi berbasis website yang dikembangkan mampu mengintegrasikan data saham, optimasi portofolio, dan estimasi Value at Risk untuk memberikan informasi investasi yang lebih akurat.
==================================================================================================================================
The capital market in Indonesia shows significant growth, where investors tend to look for stocks with strong fundamentals such as the IDX Quality 30 index, but even though the index is said to be good, market risk still exists, requiring portfolio diversification with the Markowitz model. While the Markowitz model helps with portfolio optimization, its limitations in measuring extreme risks, such as the “heavy tail” phenomenon, require a more accurate approach. Therefore, this study uses the calculation of Value at Risk with the Extreme Value Theory (EVT) method, specifically Generalized Extreme Value (GEV) and Generalized Pareto Distribution (GPD). methods, so that the estimation results are more accurate. The results of this study are expected to improve understanding of portfolio optimization and extreme risk management, while providing information for investors in making investment decisions. To support more accurate and effective investment decision-making, a website-based application was designed that offers functionality to integrate stock data, portfolio optimization, and interactive risk estimation. The results show that the optimal portfolio consists of 24% ADRO shares, 23% BNGA shares, 21% NISP shares, 16% AMRT shares, 14% BRIS shares, and 2% MIKA shares. Meanwhile, the best risk estimate is obtained using the Generalized Extreme Value (GEV) distribution, with VaR values at confidence levels of 90%, 95%, and 99% at -2.3769%, -2.5652%, and 2.8224%, respectively. The developed web-based application is capable of integrating stock data, portfolio optimization, and Value at Risk estimation to provide more accurate investment information.
Actions (login required)
 |
View Item |