Optimasi Portofolio Black-litterman Dengan Pendekatan Dependensi Clayton Copula

Budiarto, Yudhistira Rizky (2025) Optimasi Portofolio Black-litterman Dengan Pendekatan Dependensi Clayton Copula. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5006211107-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Optimasi portofolio bertujuan menyeimbangkan risiko dan return. Model Black-Litterman mempertimbangkan pandangan investor namun belum menangkap ketergantungan ekstrem antar aset. Penelitian ini menerapkan pendekatan Copula Clayton untuk memodelkan struktur dependensi non-linier, khususnya pada ekor bawah distribusi. Lima jenis aset diuji yaitu saham BBCA yang akan digabung dengan saham BBRI, indeks sektoral gabungan dan perbankan, indeks pasar perbankan, Bitcoin, dan komoditas Minyak. Data harian lima tahun terakhir ditransformasi ke domain uniform [0,1], kemudian dilakukan estimasi parameter dengan Maximum Likelihood Estimation dan seleksi model menggunakan Akaike Information Criterion dan uji goodness of fit dengan Cramer von Mises. Hasil menunjukkan bahwa seluruh portofolio paling sesuai dimodelkan dengan Copula Clayton. Pendekatan Copula Black-Litterman menghasilkan expected return dan Sharpe ratio yang lebih tinggi di seluruh portofolio dibanding pendekatan Market Cap Weighted. Selain itu, estimasi VaR portofolio secara konsisten lebih rendah dibandingkan VaR aset-aset penyusunnya, menandakan efektivitas struktur pembobotan dalam menekan risiko. Temuan ini menunjukkan bahwa pendekatan Copula Black-Litterman mampu menjadi penengah antara return dan risiko, serta menghasilkan portofolio yang lebih efisien dan adaptif terhadap struktur dependensi pasar
==================================================================================================================================
Portfolio optimization aims to balance risk and return. The Black-Litterman model incorporates investor views but lacks the ability to capture extreme dependence between assets. This study applies the Clayton Copula approach to model non-linear dependency structures, particularly lower tail dependence. Five asset classes are evaluated, combining BBCA with BBRI, sectoral and banking indices, a market index for banking, Bitcoin, and oil commodities. Five years of daily data were transformed into the [0,1] uniform domain, followed by parameter estimation using Maximum Likelihood Estimation, model selection with Akaike Information Criterion, and goodness-of-fit testing via Cramer von Mises. Results show that all portfolios are best modeled using the Clayton Copula. The Copula Black-Litterman approach yields higher expected returns and Sharpe ratios across all portfolios compared to the Market Cap Weighted method. Furthermore, portfolio VaR estimates are consistently lower than the VaR of individual constituent assets, indicating effective risk reduction through optimized weighting structures. These findings suggest that the Copula Black-Litterman method effectively balances return and risk while producing more efficient portfolios that adapt to market dependency structures.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Black-Litterman, Copula Clayton, Maximum Likelihood Estimation, Optimasi Portofolio, Simulasi Monte Carlo. Black-Litterman, Clayton Copula, Maximum Likelihood Estimation, Monte Carlo Simulation, Portfolio Optimization.
Subjects:	H Social Sciences > HG Finance > HG4529.5 Portfolio management
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Yudhistira Rizky Budiarto
Date Deposited:	25 Jul 2025 07:44
Last Modified:	25 Jul 2025 07:44
URI:	http://repository.its.ac.id/id/eprint/121585

Actions (login required)

View Item