Naslov
Implications of Measuring VaR Using Historical Simulation ; An Example of Zagreb Stock Exchange Index - Crobex

Autori
Žiković, Saša

Vrsta, podvrsta i kategorija rada
Poglavlja u knjigama, znanstveni

Knjiga
Resource Allocation and Institutions: Explorations in Economics, Finance and Law

Urednik/ci
Roufagalas, John

Izdavač
Athens Institute for Education and Research (ATINER)

Grad
Atena

Godina
2006

Raspon stranica
367-389

ISBN
960-6672-01-8

Ključne riječi
Croatia, CROBEX index, Value at Risk, Historical simulation

Sažetak
In this paper the author tests the acceptability of measuring Value at Risk (VaR) with Historical simulation in Croatian financial market, specifically Zagreb stock exchange. Since its introduction in the early ’ 80, VaR methodology has become the most popular way of measuring and managing market risk. Value at Risk can be described as the worst loss that can be expected from holding a portfolio of securities, over a certain period of time, given a specified level of probability, under normal market conditions. The most common classification of VaR methods found in literature is that of: parametric VaR estimates, Historical simulation and Monte Carlo simulation. Each of these approaches for measuring VaR has its’ own advantages and disadvantages. The author in this paper focuses on Historical simulation because of its’ three main advantages over the parametric approach for measuring VaR – Historical simulation: (1) does not presume that returns are normally distributed, (2) is simple to calculate since there is no need for variance-covariance matrices and (3) can easily accommodate skewness, kurtosis and other features of empirical distribution that cause serious problems for parametric approach. Assumption of normal “ Gaussian” distribution cannot be observed in transitional and emerging economies, and as such could seriously distort the VaR figures obtained by the use of a parametric approach. The applicability of historical simulation is tested on CROBEX index – the stock index of Zagreb stock exchange. Historical simulation with five different observation periods (50, 100, 175, 250 and 400 days) is tested on CROBEX index. VaR figures at 95% and 99% confidence level are tested for each of the different opservation periods. In conclusion the results of the research are analyzed and useful suggestions for future research are given.

Izvorni jezik
Engleski

Znanstvena područja
Ekonomija

POVEZANOST RADA