Abstrak 
MvaR Normal Vs MvaR Student By Non-constant Volatility And The Long Memory Effect
Sukono, Subanar, Dedi Rosadi
Universitas Padjadjaran, Proceedings Of 4th International Conference On Mathematics And Statistics (ICOMS 2009)
Bahasa Inggris
Universitas Padjadjaran, Proceedings Of 4th International Conference On Mathematics And Statistics (ICOMS 2009)
ARFIMA, asymmetry distribution, Fat tails, GARCH, MVaR
Objective of the paper is to investigate the comparison between the modified Value-atRisk (MVaR) by the standard normal distribution and MVaR by the standard Student-t distribution approaches, especially when the volatility is non-constant then considering the effect of long memory. In MVaR, to calculate the risk rate of investment in addition to involving the average and standard deviation, the skewness and kurtosis also must be considered, especially for the return assets with non-normal distributed. Furthermore, empirically it will be compared MVaR by the standard normal distribution approach and MVaR by the standard Student-t distribution approach, to see the ability of each approach in capturing the effect of non-normality of return assets distribution.