| Abstract [eng] |
In this master's thesis, cryptocurrency portfolio risk is assessed by combining ARMA-GJR-GARCH models and regular vine copulas. First, an ARMA(p,q)-GJR-GARCH(1,1) model is estimated for each cryptocurrency and standartized residuals are obtained. Next, the dependence structure among these residuals is modeled using regular vine copulas. Based on the fitted joint model, 10000 portfolio return simulations are generated, from which the portfolio Value at Risk is computed. Finally, the reliability of the model's forecasts is evaluated using backtesting. |
