Υπολογισμός του Value-at-Risk με τη χρήση γενικευμένων αυτοπαλίνδρομων υπό συνθήκη ετεροσκεδαστικότητας υποδειγμάτων
Μπλούτσος, Κωνσταντίνος Α.
This thesis presents the concept of Risk as it is obtained by the Value-at-Risk (VaR) method. This technique provides the user a measure of risk which is expressed by a single number for the worst expected loss of an asset for a given horizon at a given confidence level. VaR is estimated by the aid of a sophisticated econometric method which combines time series analysis (ARIMA analysis) and generalized autoregressive conditional heteroskedastic (GARCH) models. Using these models, VaR is produced as one-step-ahead forecast of the process. This method is applied to log returns of stock prices of European Telecommunication Companies in order to determine risk not only on an individual basis, but also in a marketwise basis.