Czasopismo
Tytuł artykułu
Autorzy
Warianty tytułu
Języki publikacji
Abstrakty
Financial risk management is very important activity of many Financial institutions and enterprises. Due to the increasing risk in the Financial markets, this activity has crucial impact on the value of the institution. Therefore, more and more efforts are undertaken to design the effective system of Financial risk management. The key stage in risk management is risk measurement. The crucial role in risk management is played by the relationship between re- tums or between the changes of the prices. Unfortunately, most theoretical results are derived in the case of linear relationship. This occurs when multivariate distribution of the returns is elliptically symmetric distribution, including multivariate normal distribution. Here, the correlation coefficient is the proper measure of the relationship. However, the distribution of returns on the financial markets very often significantly departs from elliptically symmetric distribution. In such cases, using correlation coefficient (or correlation matrix) as a theoretical tool for risk management is not justifiable. This refers also to extreme dependence. This is the dependence between extreme values of two variables, for example two returns. In this paper we present the different approach, which is more general and avoids the drawbacks of correlation coefficient. It is called copula analysis. We will show how this approach can be used to measure extreme dependence. (fragment of text)
Rocznik
Strony
179--184
Opis fizyczny
Twórcy
autor
- Wrocław University of Economics, Poland
autor
- Wrocław University of Economics, Poland
Bibliografia
- Nelsen R. (1999): An introduction to copulas. New York: Springer-Verlag.
- Sklar A. (1959): Fonctions de repartition a n dimensions et leurs marges. Publications de 1'Institut de Statistique de l'Universite de Paris 8, pp.229-231.
Typ dokumentu
Bibliografia
Identyfikatory
Identyfikator YADDA
bwmeta1.element.ekon-element-000171460372