Predikce kurzů akcií s využitím modelu value at risk

Abstract

Value at Risk model is often used for risk analyses mostly in the banking and insurance industries. Following the characteristics of the model principle, the Value at Risk is interpreted in the economic sense. Attention is paid to three sub-methods, concretely historical simulation method, variance covariance method and Monte Carlo method. A number of empirical studies focused on the application of these submethods in practice is presented. Various risk factors are used by these studies. Value at Risk model is applied to selected stocks from SPAD segment of Prague Stock Exchange within the 2011, in the paper. This unique application is the aim of the paper. The reliability interval, hold period, historical period and other important parameters related to the sub-methods are selected within the application. Using covariance matrices, correlation matrices as well as other types of matrices and statistical indicators, the Value at Risk are calculated. The comparison of calculated diversified and non-diversified Value at Risk by sub-methods is realized. Mentioned are also back testing, stress testing, the essence of the relative and marginal Value at Risk and other options of practical application of this model.