Praja, Asta Destian Yudha (2026) Analisis Perbandingan Risiko Cryptocurrency Bitcoin, Binance, dan Ripple Berdasarkan Conditional Value at Risk dengan Pendekatan Metode ARIMA-GARCH. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5003221175-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5003221175-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (10MB) | Request a copy |
Abstract
Cryptocurrency merupakan instrumen investasi digital yang menawarkan potensi keuntungan tinggi, tetapi memiliki volatilitas yang besar sehingga diperlukan pengukuran risiko yang memadai. Penelitian ini bertujuan memodelkan karakteristik return dan volatilitas bersyarat, mengestimasi risiko individual menggunakan Value at Risk (VaR), serta menganalisis transmisi risiko antar-cryptocurrency menggunakan Conditional Value at Risk (CoVaR) dan ΔCoVaR. Data yang digunakan berupa harga penutupan harian Bitcoin (BTC), Binance Coin (BNB), dan Ripple (XRP) selama periode 1 Januari 2024 hingga 31 Desember 2025 yang ditransformasikan menjadi log return. Pemodelan rata-rata dan volatilitas dilakukan menggunakan pendekatan ARIMA–GARCH, sedangkan VaR diestimasi pada tingkat kepercayaan 95% dan 99% serta dievaluasi menggunakan uji Kupiec. Estimasi CoVaR dilakukan melalui regresi kuantil pada kuantil 0,05 dan 0,01. Hasil penelitian menunjukkan bahwa model terbaik adalah ARIMA(1,0,0)–GARCH(1,1) untuk Bitcoin, ARIMA(2,0,2)–GARCH(1,1) untuk Binance, dan ARIMA(1,0,1)–GARCH(1,2) untuk Ripple. Ketiga aset menunjukkan pola volatility clustering, dengan persistensi volatilitas tertinggi pada Bitcoin, sedangkan Ripple paling responsif terhadap shock baru. Berdasarkan nilai absolut rata-rata VaR, Ripple memiliki risiko kerugian harian tertinggi, yaitu sebesar 6,1398% pada tingkat kepercayaan 95% dan 11,7565% pada tingkat kepercayaan 99%. Hasil backtesting menunjukkan bahwa seluruh model VaR valid karena uji Kupiec menghasilkan p-value lebih besar dari 0,05. Analisis CoVaR dan ΔCoVaR menunjukkan bahwa transmisi risiko antar-aset bersifat tidak simetris. Pada tingkat kepercayaan 99%, risiko bersyarat terbesar terjadi pada Ripple ketika Bitcoin berada dalam kondisi distress, dengan mean CoVaR sebesar -15,9240% dan mean ΔCoVaR sebesar -6,8923%. Temuan ini menunjukkan bahwa pengelolaan portofolio cryptocurrency perlu mempertimbangkan tidak hanya risiko individual, tetapi juga arah dan besarnya transmisi risiko antar-aset.
=====================================================================================================================================
Cryptocurrency is a digital investment instrument that offers high return potential but is also characterized by substantial volatility, making adequate risk measurement essential. This study aims to model the return characteristics and conditional volatility, estimate individual risk using Value at Risk (VaR), and analyze risk transmission among cryptocurrencies using Conditional Value at Risk (CoVaR) and ΔCoVaR. The data consist of daily closing prices of Bitcoin (BTC), Binance Coin (BNB), and Ripple (XRP) from January 1, 2024, to December 31, 2025, which were transformed into log returns. The mean and volatility processes were modeled using the ARIMA–GARCH approach, while VaR was estimated at the 95% and 99% confidence levels and evaluated using the Kupiec test. CoVaR was estimated through quantile regression at the 0.05 and 0.01 quantiles. The results show that the best models are ARIMA(1,0,0)–GARCH(1,1) for Bitcoin, ARIMA(2,0,2)–GARCH(1,1) for Binance Coin, and ARIMA(1,0,1)–GARCH(1,2) for Ripple. All three assets exhibit volatility clustering, with Bitcoin showing the highest volatility persistence, while Ripple is the most responsive to new shocks. Based on the absolute mean VaR, Ripple has the highest daily loss risk, amounting to 6.1398% at the 95% confidence level and 11.7565% at the 99% confidence level. The backtesting results indicate that all VaR models are valid because the Kupiec test produces p-values greater than 0.05. The CoVaR and ΔCoVaR analyses show that risk transmission among the assets is asymmetric. At the 99% confidence level, the highest conditional risk occurs in Ripple when Bitcoin is under distress, with a mean CoVaR of -15.9240% and a mean ΔCoVaR of -6.8923%. These findings indicate that cryptocurrency portfolio management should consider not only the individual risk of each asset but also the direction and magnitude of risk transmission among assets.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | ARIMA-GARCH, Volatilitas, Cryptocurrency, Conditonal Value at Risk, ARIMA-GARCH, Volatility, Cryptocurrency, Conditional Value at Risk |
| Subjects: | H Social Sciences > HA Statistics > HA30.3 Time-series analysis H Social Sciences > HA Statistics > HA31.7 Estimation H Social Sciences > HG Finance > HG4529 Investment analysis |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Asta Destian Yudha Praja |
| Date Deposited: | 04 Aug 2026 07:35 |
| Last Modified: | 04 Aug 2026 07:35 |
| URI: | http://repository.its.ac.id/id/eprint/142474 |
Actions (login required)
 |
View Item |