AN APPLICATION ON VALUE AT RISK ANALYSIS: EVIDENCE FROM TURKEY

Determining the maximum loss that occurs in the value of a portfolio in a given period of time has a great importance in terms of firms and investors. Risk management has become a serious issue with the rapid growth in financial markets and globalization phenomena. One of the most important risk measurement methods developed in recent years is the Value at Risk (VaR) method. In this study, a hypothetical portfolio amounting to 100,000 TL consisting of the shares of 5 companies in the BIST 30 index was analyzed by Parametric, Historical Simulation and Monte Carlo Simulation methods from VaR techniques based on constant variance for 2015, 2016 and 2017 years. According to the results, the smallest amount of VaR is obtained by the Historical Simulation method which does not have any assumptions about the statistical distributions of the series and based on past conditions to be valid in the future. Z Test which is used for backtesting indicates that the parametric method has been appropriate for all three years, while the Mixed Kupiec Test, which takes into account the dependence between the deviation times, reveals that the method is not suitable only for 2017