The comparison of Value at Risk (VaR) and Conditional Value at Risk (CVaR) model : The case of Cryptocurrency

Abstract

This study is on comparing Value at Risk (VaR) and Conditional Value at Risk (CVaR) model, which include GARCH and Extreme Value Theory (EVT) and Historical Simulation to calculate the market risk of market index and Cryptocurrency. Consists of Set Index (SET), Nasdaq Composite (IXIC), Cryptocurrency Index (CRIX), Bitcoin (BTC), Ethereum (ETH) and Binance Coin (BNB). Investor are able to use Value at Risk (VaR) and Conditional Value at Risk (CVaR) as a tool to foresee the possible loss incurred due to return movement and adjust portfolio to be under an acceptable level of risk. The framework of Value at Risk (VaR) and Conditional Value at Risk (CVaR) model is calculating at 95% and 99% confidence level and also back-testing in which Violation Ratio and Three-zone Approach are applied.

The result shows that GARCH-EVT model is more suitable and pass back testing to estimate Value at Risk (VaR) and Conditional Value at Risk (CVaR) of market index and Cryptocurrency both 95% and 99% confidence level. Besides, the study finds that Binance coin (BNB) and Ethereum (ETH) presents the highest estimated risk, where as it provides the highest return, therefore it is attractive to risk-loving investor.