Rani, Inda Usha (2025) Estimasi Risiko Investasi Cryptocurrency dengan Model Hybrid Generalized Autoregressive Conditional Heteroscedasticity - Extreme Value Theory. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5006211060-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5006211060-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (8MB) | Request a copy |
Abstract
Kripto telah menjadi salah satu aset digital dengan pertumbuhan yang sangat pesat dalam beberapa tahun terakhir. Di Indonesia, aset kripto dapat digunakan sebagai alat investasi berupa komoditi yang diperdagangkan di bursa berjangka. Namun, volatilitas harga yang ekstrem menjadikan investasi ini berisiko tinggi. Oleh karena itu, diperlukan metode yang dapat mengukur risiko tersebut, salah satunya menggunakan pendekatan Value at Risk (VaR) dan Expected Shortfall (ES). Penelitian ini mengusulkan pendekatan hybrid Generalized Autoregressive Conditional Heteroskedasticity - Extreme Value Theory (GARCH-EVT) untuk mengestimasi risiko investasi kripto. Model GARCH digunakan untuk menangkap volatilitas yang bersifat heteroskedastis, sedangkan EVT digunakan untuk mengatasi distribusi ekor ekstrem. Penelitian ini menggunakan data harga penutupan harian dari lima aset kripto dengan kapitalisasi pasar tertinggi, yaitu Bitcoin, Ethereum, Tether USDt, Binance Coin, dan Solana, dalam periode 1 Januari 2018 hingga 31 Desember 2024. Hasil penelitian menunjukkan bahwa pada tingkat kepercayaan 95% dan 99% nilai risiko kerugian tertinggi dimiliki oleh kripto Solana dan nilai risiko kerugian terendah dimiliki oleh kripto Tether USDT. Berdasarkan hasil backtesting, pendekatan GARCH-EVT memberikan estimasi VaR yang akurat untuk kripto Tether USDt pada tingkat kepercayaan 95%, serta untuk Tether USDT, Bitcoin, dan Binance pada tingkat kepercayaan 99%. Namun, estimasi VaR untuk kripto lainnya dinyatakan tidak akurat pada sebagian besar tingkat kepercayaan. Sebagai ukuran risiko alternatif, nilai ES dapat digunakan untuk mengetahui ekspektasi nilai kerugian di atas VaR.
======================================================================================================================================
Crypto has become one of the fastest-growing digital assets in recent years. In Indonesia, crypto assets can be used as an investment tool in the form of commodities traded in future exchanges. However, the extreme price volatility makes this investment highly risk. Therefore, a method that can measure this risk is needed, such as Value at Risk (VaR) and Expected Shortfall (ES) approaches. This study proposes a hybrid Generalized Autoregressive Conditional Heteroskedasticity - Extreme Value Theory (GARCH-EVT) approach to estimate crypto investment risk. The GARCH model is used to capture heteroskedastic volatility, while EVT is used to address extreme tail distribution. This study uses daily closing price data of five crypto assets with the highest market capitalization, namely Bitcoin, Ethereum, Tether USDT, Binance Coin, and Solana, from January 1, 2018, to December 31, 2024. The results show that at both 95% and 99% confidence levels, Solana cryptocurrency had the largest chance of loss, whereas Tether USDt cryptocurrency had the lowest risk. Based on backtesting result, the GARCH-EVT approach provides accurate VaR estimations for Tether USDT at a 95% confidence level, and for Tether USDt, Bitcoin, and Binance Coin at a 99% confidence level. However, VaR estimations for other cryptocurrencies were found to be inaccurate across most confidence levels. As an alternative risk measure, the ES value can be used to determine the expected loss value exceeding VaR.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Cryptocurrency, Expected Shortfall, Extreme Value Theory, GARCH, 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 Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis |
| Depositing User: | Inda Usha Rani |
| Date Deposited: | 28 Jul 2025 10:32 |
| Last Modified: | 28 Jul 2025 10:32 |
| URI: | http://repository.its.ac.id/id/eprint/122384 |
Actions (login required)
 |
View Item |