<![CDATA[Stock is one of investments that used by investor but often have high risk. So we need to calculate risk assessment for single stock and portfolios. Value at Risk (VaR) is a tool often used in measuring risk, especially in stock trading. Return stock usually has a fat tail distribution, there is usually a case of heteroscedasticity. Time series model that used to modeling this condition is Autoregressive Conditional Heteroscedasticity / Generalized Autoregressive Conditional Heteroscedasticity. This study focused on the calculation of VaR using Block Maxima with the approach Generalized Extreme Value/GEV and Peaks Over Threshold approach Generalized Pareto Distribution/GPD. Modeling volatility models of GARCH. Share data used in the case study is a daily closing PT. Astra International and Panin Financial period January 1st, 2010 – January 22nd, 2016. The result is ARIMA(0,1,1) GARCH(1,2) which is the best model with the smallest AIC. The amount of risk with a confidence level of 95% by GEV is 3,1613%, while the GPD is 3,2761% rupiah from current asset, in other words VaR GPD higher better than GEV.Keywords: Portfolio, Return, Value at Risk (VaR), ARCH/GARCH, Block Maxima, Peaks Over Threshold, GEV, GPD.]]>
Copyrights © 2016
