Implementasi Hierarchical Risk Parity, Hierarchical Clustering Based Asset Allocation, dan Hierarchical Equal Risk Contribution Pada Pembentukan Portofolio Crypto Currency

Soegiarto, Galvan Fattah (2026) Implementasi Hierarchical Risk Parity, Hierarchical Clustering Based Asset Allocation, dan Hierarchical Equal Risk Contribution Pada Pembentukan Portofolio Crypto Currency. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5006221101_Undergraduate Thesis.pdf] Text
5006221101_Undergraduate Thesis.pdf - Accepted Version
Restricted to Repository staff only
Download (16MB) | Request a copy
[thumbnail of 5006221101_Undergraduate Thesis.pdf] Text
5006221101_Undergraduate Thesis.pdf
Restricted to Repository staff only
Download (16MB) | Request a copy

Abstract

Investasi merupakan komitmen yang dilakukan untuk memperoleh keuntungan di masa depan. Investasi dapat dilakukan pada berbagai instrumen dan setiap instrumen memiliki tingkat risiko yang berbeda. Risiko ini berbanding lurus dengan tingkat keuntungan yang bisa diperoleh. Untuk memaksimalkan keuntungan, sebagian investor memilih crypto currency sebagai instrumen investasi dengan tingkat keuntungan dan risiko yang relatif paling tinggi di antara instrumen investasi yang lain. Akan tetapi, tanpa adanya strategi pembentukan portofolio yang tepat, investor bisa saja mengalami kerugian yang signifikan. Pasar crypto currency memiliki kondisi yang unik dan berbeda dengan instrumen lain. Bitcoin sebagai crypto currency terlama dan kapitalisasi pasar paling besar memiliki pengaruh yang signifikan terhadap pergerakan pasar crypto currency. Aset crypto currency lainnya memiliki korelasi tinggi secara positif terhadap Bitcoin. Kondisi seperti ini membuat pendekatan pembentukan portofolio dengan mean-variance optimization, sebagai salah satu metode populer, tidak bisa melakukan diversifikasi dengan baik. Oleh karena itu, penelitian ini dilakukan dengan metode pembentukan portofolio berdasarkan hierarki yang dikembangkan untuk menjawab keterbatasan metode mean-variance optimization. Untuk melihat metode hierarki mana yang paling optimal, penelitian dilakukan dengan beberapa metode pembentukan portofolio berbasis hierarki, seperti hierarchical risk parity, hierarchical clustering based aset allocation, dan hierarchical equal risk contribution. Penelitian ini juga akan membandingkan metode tersebut dengan beberapa metrik, seperti adjusted sharpe ratio, certainty-equivalent return, max drawdown, average turnover, dan sum of squared portfolio weights. Kemudian, data yang dipilih adalah dua puluh aset crypto currency dengan kapitalisasi pasar tertinggi mengingat bahwa jumlah kapitalisasi pasar yang besar cenderung memiliki likuiditas yang besar juga. Penelitian ini menunjukkan bahwa pendekatan pembobotan portofolio berbasis hierarki menghasilkan kinerja yang beragam baik dari sisi return maupun risiko. Berdasarkan return portofolio, metode hierarchical equal risk contribution (HERC) dengan single linkage dan metrik risiko CDaR menghasilkan return tertinggi sebesar 46,08%, diikuti oleh metode hierarchical clustering–based asset allocation (HCAA) dengan Ward linkage sebesar 27,21%, sementara metode hierarchical risk parity (HRP) hanya menghasilkan return sebesar 0,03%. Dari sisi risiko, metode HERC menunjukkan keunggulan dalam mengakomodasi faktor risiko dan mengendalikan kerugian ekstrem, sedangkan HCAA memiliki nilai adjusted Sharpe ratio tertinggi sebesar 0,1603 dan tingkat diversifikasi yang baik. Di sisi lain, metode HRP menghasilkan maximum drawdown terendah sebesar −0,10% serta nilai turnover nol, yang menunjukkan intensitas rebalancing yang sangat rendah. Temuan ini mengindikasikan bahwa metode pembentukan portofolio berbasis hierarki merupakan alternatif yang relevan dan efektif untuk pasar crypto currency yang memiliki korelasi tinggi dan karakteristik risiko ekstrem.
======================================================================================================================================
Investment is a commitment made to gain returns in the future. Investments can be made in various instruments, each with different levels of risk. This risk is directly proportional to the potential return that can be obtained. To maximize profits, some investors choose cryptocurrency as an investment instrument, given its relatively high returns and risks compared to other investment options. However, without an appropriate portfolio construction strategy, investors may face significant losses. The cryptocurrency market has unique characteristics that differ from other instruments. Bitcoin, as the oldest cryptocurrency and the one with the largest market capitalization, has a significant influence on the overall cryptocurrency market movement. Other cryptocurrencies are highly positively correlated with Bitcoin. This condition makes portfolio construction using the mean-variance optimization method, one of the most popular approaches, less effective in achieving proper diversification. Therefore, this study employs hierarchical-based portfolio construction methods, which are designed to address the limitations of mean-variance optimization. To determine which hierarchical method is the most optimal, the study applies several hierarchical portfolio construction techniques, such as hierarchical risk parity (HRP), hierarchical clustering-based asset allocation (HCAA), and hierarchical equal risk contribution (HERC). These methods are evaluated using several metrics, including adjusted sharpe ratio, certainty-equivalent return, max drawdown, average turnover, and sum of squared portfolio weights. The study focuses on twenty cryptocurrencies with the largest market capitalizations, considering that assets with high market capitalization tend to have higher liquidity as well. This study shows that hierarchical portfolio weighting approaches exhibit heterogeneous performance in terms of both return and risk. In terms of portfolio return, the hierarchical equal risk contribution (HERC) method with single linkage and CDaR risk metric achieves the highest return of 46.08%, followed by the hierarchical clustering–based asset allocation (HCAA) method with Ward linkage at 27.21%, while the hierarchical risk parity (HRP) method records a substantially lower return of 0.03%. From a risk perspective, HERC demonstrates superior capability in incorporating risk factors and controlling extreme losses, as reflected in its favorable certainty-equivalent return and diversification measures. Meanwhile, HCAA achieves the highest adjusted Sharpe ratio of 0.1603 and exhibits strong diversification properties, whereas HRP attains the lowest maximum drawdown of −0.10% and zero portfolio turnover, indicating minimal rebalancing activity. These findings indicate that hierarchical portfolio construction methods provide an effective alternative for cryptocurrency markets characterized by high correlations and extreme risk dynamics.

Item Type: Thesis (Other)
Uncontrolled Keywords: Crypto Currency, Hierarchical Clustering Based Aset Allocation, Hierarchical Equal Risk Contributiin, Hierarchical Risk Parity, Manajemen Portofolio. Crypto Currency, Hierarchical Clustering Based Aset Allocation, Hierarchical Equal Risk Contribution, Hierarchical Risk Parity, Portfolio Management.
Subjects: Q Science > QA Mathematics > QA278.55 Cluster analysis
Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science)
Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis. Correspondence analysis (Statistics)
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User: Galvan Fattah Soegiarto
Date Deposited: 13 Jan 2026 05:49
Last Modified: 13 Jan 2026 05:49
URI: http://repository.its.ac.id/id/eprint/129553

Actions (login required)

View Item View Item