Application of VaR (Value at Risk) method on Belgrade Stock Exchange (BSE) optimal portfolio

Miletić, Siniša and Korenak, Boris and Lutovac, Miloš (2014) Application of VaR (Value at Risk) method on Belgrade Stock Exchange (BSE) optimal portfolio. International review (1‐2). pp. 142-159. ISSN 2217-9739

[img] Text
Application of VaR (Value at Risk) method on Belgrade Stock Exchange (BSE) optimal portfolio .pdf - Published Version
Restricted to Repository staff only
Download (703kB)

Abstract

The main objective of this study is to determine the adequacy of the measurement of market risks of financial institutions in Serbia by the method of Value at Risk(VaR. For investors, in the current global financial crisis, it is particularly important to accurately measure and allocate risk and efficiently manage their portfolio. Possibility of application of VaR methodology, which is basically designed and developed for liquid and developed markets, should be tested on the emerging markets, which are characterized by volatility, illiquidity and shallowness of the market. Value of VaR in this study was calculated using historical and parametric methods and backtesting analysis was used to verify the adequacy of the application of VaR models. Backtesting VaR model performance analysis was conducted to compare the ex-ante VaR estimate to the ex-post returns. The empirical results show that parameter exponentially weighted moving average model gives lower values at risk in both cases (95% and 99%) due to the fact that this method assigns weights to more recent returns while our portfolio is exposed to a lower volatility in recent time. Based on the results of Kupiec's and Christoffersen's test, it was observed that VaR estimates obtained by both, parametric and historical simulation, give a good prediction of market risk, at 95% and 99% confidence level.

Item Type: Article
Uncontrolled Keywords: market risk, optimal portfolio, Value at Risk (VaR), backtesting, EWMA, GARCH models, historical simulation
Subjects: 300 Scienze sociali
Depositing User: Marina Spanti
Date Deposited: 25 Jan 2016 15:05
Last Modified: 25 Jan 2016 15:05
URI: http://eprints.bice.rm.cnr.it/id/eprint/10374

Actions (login required)

View Item View Item
CNR Solar is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.