COMPARATIVE ANALYSIS OF COVARIANCE AND OPTIMIZATION METHODS FOR ESTIMATING RISKS FOR THE PORTFOLIO WITH PAPERS.

In this work two Value-at-Risk (VaR) estimation models are compared: covariance method that is commonly used in Ukrainian banks and optimization one suggested by Uryasev and Rockafellar. Conditional Value-at-Risk (CVaR) that is an alternative risk measure to VaR, that permits to estimate the expected loss beyond VaR, is considered. Our work extends the Uryasev-Rockafellar approach to the quadratic portfolio containing derivative financial instruments. VaR estimation model's behaviour is examined as based on the hypothesis that the risk factors have elliptical distribution. [ABSTRACT FROM AUTHOR]