Analisis Dependensi Volatilitas Antara Cryptocurrency Dan Indeks Harga Saham Gabungan Dengan Pendekatan Dynamic Conditional Correlation-GARCH-Copula

Hafizhah, Anindya Zhafira Noer (2025) Analisis Dependensi Volatilitas Antara Cryptocurrency Dan Indeks Harga Saham Gabungan Dengan Pendekatan Dynamic Conditional Correlation-GARCH-Copula. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5006211013-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (7MB) | Request a copy

Abstract

Popularitas cryptocurrency sebagai instrumen investasi telah mengalami peningkatan pesat dalam beberapa tahun terakhir. Namun, tingginya volatilitas pada aset digital seperti Bitcoin (BTC) dan Ethereum (ETH) menjadikannya instrumen dengan risiko tinggi, sehingga penting untuk memahami perilaku dan interaksi dengan pasar keuangan tradisional seperti Indeks Harga Saham Gabungan (IHSG). Penelitian ini bertujuan untuk menganalisis hubungan ketergantungan volatilitas antara aset kripto dan pasar saham Indonesia, khususnya antara Bitcoin (BTC), Ethereum (ETH), dan Indeks Harga Saham Gabungan (IHSG), dengan menggunakan pendekatan Dynamic Conditional Correlation-Generalized Autoregressive Conditional Heteroskedasticity dan Copula (DCC-GARCH-Copula). Data yang digunakan meliputi harga penutupan harian dari ketiga aset selama periode 1 Januari 2022 hingga 31 Desember 2024. Proses analisis diawali dengan identifikasi model ARIMA dan GARCH(1,1) secara univariat, dilanjutkan dengan estimasi model DCC-GARCH untuk memperoleh korelasi dinamis, dan diakhiri dengan pemodelan struktur dependensi non-linear menggunakan copula, khususnya Gaussian dan t-copula untuk memperoleh tail dependence. Hasil penelitian menunjukkan bahwa volatilitas harga pada BTC dan ETH jauh lebih tinggi dibandingkan dengan IHSG, baik dari segi rentang nilai maupun standar deviasi. Model DCC-GARCH berhasil menangkap dinamika korelasi waktu-ke-waktu, dengan korelasi rata-rata antara IHSG-BTC dan IHSG-ETH masing-masing sebesar 0,0325 dan 0,0549, sedangkan korelasi antara BTC-ETH mencapai 0,8399. Selain itu, estimasi menggunakan t-copula menghasilkan nilai log-likelihood yang lebih tinggi dibandingkan Gaussian copula pada semua pasangan aset, mengindikasikan bahwa t-copula lebih tepat dalam menangkap hubungan non-linear dan tail dependence. Nilai tail dependence tertinggi terdapat pada pasangan BTC-ETH sebesar 0,3798, sedangkan pada pasangan IHSG-BTC dan IHSG-ETH, nilai tersebut mendekati nol. Temuan ini menunjukkan bahwa keterkaitan volatilitas antara pasar saham Indonesia dan pasar kripto relatif lemah, namun terdapat hubungan ekstrem yang kuat antar sesama aset kripto. Oleh karena itu, aset kripto dapat dipertimbangkan sebagai alternatif diversifikasi portofolio, namun tetap memerlukan perhatian khusus terhadap risiko ekstrem di antara aset kripto itu sendiri.
============================================================================================================================
The popularity of cryptocurrency as an investment instrument has surged significantly in recent years. However, the high volatility of digital assets such as Bitcoin (BTC) and Ethereum (ETH) renders them high-risk instruments, making it essential to understand their behavior and interactions with traditional financial markets, such as the Indonesia Stock Exchange Composite Index (IHSG). This study aims to analyze the volatility dependence between cryptocurrencies and the Indonesian stock market, particularly between Bitcoin (BTC), Ethereum (ETH), and IHSG, using the Dynamic Conditional Correlation-Generalized Autoregressive Conditional Heteroskedasticity and Copula (DCC-GARCH-Copula) approach. The dataset consists of daily closing prices of the three assets from January 1, 2022, to December 31, 2024. The analysis begins with the identification of univariate ARIMA and GARCH(1,1) models, followed by the estimation of the DCC-GARCH model to obtain dynamic correlations over time, and concludes with the modeling of the non-linear dependency structure using Gaussian and t-copula functions. The results show that the price volatility of BTC and ETH is significantly higher than that of IHSG, both in terms of value range and standard deviation. The DCC-GARCH model successfully captures time-varying correlations, with average correlations of 0.0325 and 0.0549 for IHSG-BTC and IHSG-ETH respectively, and a much higher correlation of 0.8399 between BTC and ETH. Furthermore, the t-copula provides a higher log-likelihood than the Gaussian copula across all asset pairs, indicating its superior ability to capture non-linear and tail dependencies. The highest tail dependence value is observed in the BTC-ETH pair (0.3798), whereas IHSG-BTC and IHSG-ETH exhibit near-zero tail dependence values. These findings indicate that the volatility linkage between the Indonesian stock market and cryptocurrencies is relatively weak, while there is a strong tail dependence between the crypto assets themselves. Therefore, cryptocurrencies may be considered as alternative portfolio diversification instruments, yet require careful attention to extreme risk spillovers within the crypto market.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Bitcoin (BTC), Cryptocurrency, DCC-GARCH-Copula, IHSG
Subjects:	H Social Sciences > HA Statistics > HA30.3 Time-series analysis
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Anindya Zhafira Noer Hafizhah
Date Deposited:	28 Jul 2025 02:29
Last Modified:	28 Jul 2025 02:29
URI:	http://repository.its.ac.id/id/eprint/121898

Actions (login required)

View Item