摘要：VaR is used as a standard tool to measure the potential loss in value of risky asset or portfolio. Calculation of VaR is commonly based on the assumption that the log return of asset or portfolio is normally dis-tributed. In this paper we calculate the VaR by assuming the log return of asset or portfolio follows the VG process. The goal of our work it to apply a bivariate VG process via copulas technique by using property proposed by Luciano and Schoutensthat the VG process conditioning on a realization of the stochastic Gamma time change is lognormally distributed. By applying that property we calculate the VaR of a portfolio of several assets under the assumption that the log returmof each asset in a portfolio follows the VG process.
World Congress on Engineering and Technology（CET 2015）
宏观经济管理与可持续发展; 金融; 投资