Analisis Perbandingan Optimasi Portofolio Pada Black Swan Events Menggunakan Mean-Variance Optimization Dan Mean-Conditional Value At Risk

Putera, I Gede Nyoman Dharma (2025) Analisis Perbandingan Optimasi Portofolio Pada Black Swan Events Menggunakan Mean-Variance Optimization Dan Mean-Conditional Value At Risk. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5006211069-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (7MB) | Request a copy

Abstract

Pasar keuangan global kerap mengalami peristiwa tak terduga dengan dampak signifikan yang dikenal sebagai Black Swan Events, seperti krisis finansial 2008 dan pandemi COVID-19. Peristiwa ini menimbulkan volatilitas tinggi dan distribusi return yang heavy-tailed, sehingga menuntut metode pengelolaan risiko dan optimasi portofolio yang lebih adaptif. Penelitian ini bertujuan untuk membandingkan efektivitas metode Mean-Variance Optimization (MVO) dan Mean-Conditional Value at Risk (Mean-CVaR) dalam membentuk portofolio optimal pada dua periode Black Swan Events tersebut. Portofolio dibentuk berdasarkan data historis harga penutupan harian dari sepuluh saham lintas sektor, yaitu BBCA, BBRI, BUMI, CPIN, CTRA, INDF, PGAS, PWON, SMDR, dan TMAS, dengan evaluasi menggunakan Value at Risk (VaR) dan expected return. Pada periode krisis finansial 2008, metode MVO menghasilkan portofolio dengan bobot tertinggi pada saham CPIN 45,14%, PWON 26,05%, INDF 14,54%, BUMI 7,34%, TMAS 5,55%, dan BBCA 1,37%, dengan expected return sebesar 0,63%, VaR sebesar -4,14%. Sementara pada periode pandemi COVID-19, MVO memberikan portofolio dengan distribusi yang relatif merata antara TMAS 34,35%, BBCA 33,29%, dan SMDR (32,36%), menghasilkan expected return sebesar 0,23%, VaR sebesar -2,69%. Sebaliknya, metode Mean-CVaR Optimization pada periode krisis 2008 menghasilkan portofolio yang lebih terkonsentrasi pada saham CPIN 51,28%, TMAS 16,04%, BUMI 12,75%, PWON 11,95%, dan INDF 7,97%, dengan expected return sebesar 1,24%, VaR sebesar -6,08%. Sedangkan pada periode pandemi COVID-19, MCVaR menghasilkan portofolio dengan bobot dominan pada TMAS 63,62%, SMDR 22,03%, dan BBCA 14,35%, dengan expected return sebesar 0,31%, VaR sebesar -4,04%. Metode Mean-CVaR memberikan expected return yang lebih tinggi dibandingkan MVO, namun disertai dengan standar deviasi yang lebih besar. Meskipun demikian, standar deviasi yang lebih tinggi tersebut mencerminkan sensitivitas MCVaR terhadap risiko ekstrem (tail risk) dan memenuhi sifat koheren subaditivitas, sehingga metode ini dinilai lebih mampu menangkap dan mengelola potensi kerugian besar pada kondisi pasar yang bergejolak. Temuan ini menegaskan pentingnya pemilihan metode optimasi yang sesuai dengan karakteristik risiko ekstrem dan mendukung perlunya strategi diversifikasi sektor dalam Menyusun portofolio saat terjadi Black Swan Events.
=========================================================================================================================================
The global financial market frequently encounters unpredictable events with significant impacts, known as Black Swan Events, such as the 2008 financial crisis and the COVID-19 pandemic. These events trigger high volatility and heavy-tailed return distributions, thus requiring more adaptive portfolio optimization and risk management methods. This study aims to compare the effectiveness of the Mean-Variance Optimization (MVO) and Mean-Conditional Value at Risk (Mean-CVaR) methods in forming optimal portfolios during the two Black Swan periods. The portfolios were constructed using historical daily closing prices from ten cross-sector stocks, namely BBCA, BBRI, BUMI, CPIN, CTRA, INDF, PGAS, PWON, SMDR, and TMAS, and were evaluated based on Value at Risk (VaR) and expected return. In the 2008 financial crisis period, the MVO method produced a portfolio with the highest weights allocated to CPIN 45.14%, PWON 26.05%, INDF 14.54%, BUMI 7.34%, TMAS 5.55%, and BBCA 1.37%, yielding an expected return of 0.63% and a VaR of -4.14%. Meanwhile, during the COVID-19 pandemic period, MVO resulted in a more evenly distributed portfolio between TMAS 34.35%, BBCA 33.29%, and SMDR 32.36%, with an expected return of 0.23% and a VaR of -2.69%. In contrast, the Mean-CVaR Optimization method during the 2008 crisis generated a more concentrated portfolio on CPIN 51.28%, TMAS 16.04%, BUMI 12.75%, PWON 11.95%, and INDF 7.97%, with an expected return of 1.24% and a VaR of -6.08%. During the COVID-19 period, the Mean-CVaR method resulted in a portfolio heavily weighted towards TMAS 63.62%, SMDR 22.03%, and BBCA 14.35%, producing an expected return of 0.31% and a VaR of -4.04%. The Mean-CVaR method delivered a higher expected return compared to MVO but with a greater standard deviation. Nevertheless, this higher standard deviation reflects Mean-CVaR's greater sensitivity to extreme (tail) risks and its compliance with the coherent risk property of subadditivity, making it more effective in capturing and managing potential large losses during turbulent markets. These findings emphasize the importance of selecting an optimization method aligned with the characteristics of extreme risk and support the need for sector diversification strategies when constructing portfolios during Black Swan Events.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Black Swan Events, Mean-Variance Optimization, Mean-Conditional Value at Risk, Value at Risk, Conditional Value at Risk, Black Swan Events, Mean-Variance Optimization, Mean-Conditional Value at Risk, Value at Risk, Conditional Value at Risk.
Subjects:	Q Science > QA Mathematics > QA184 Algebra, Linear Q Science > QA Mathematics > QA275 Theory of errors. Least squares. Including statistical inference. Error analysis (Mathematics) Q Science > QA Mathematics > QA401 Mathematical models. Q Science > QA Mathematics > QA402.5 Genetic algorithms. Interior-point methods. Q Science > QA Mathematics > QA76.9 Computer algorithms. Virtual Reality. Computer simulation.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	I Gede Nyoman Dharma Putera
Date Deposited:	30 Jul 2025 04:04
Last Modified:	30 Jul 2025 04:04
URI:	http://repository.its.ac.id/id/eprint/123455

Actions (login required)

View Item