Abstrak 
Optimisasi Portofolio Mean-VaR Dengan Volatilitas Tak Konstan Dan Efek Long Memory
Sukono, Subanar, Dedi Rosadi
Universitas Padjadjaran, Prosiding Seminar Nasional Matematika 2010 " Matematika Dalam Riset, Teknologi dan Pendidikan Surakarta, 7 Agustus 2010 ISBN: 979-978-3456-79-1
Bahasa Indonesia, Bahasa Inggris
Universitas Padjadjaran, Prosiding Seminar Nasional Matematika 2010 " Matematika Dalam Riset, Teknologi dan Pendidikan Surakarta, 7 Agustus 2010 ISBN: 979-978-3456-79-1
ARFIMA, GARCH, Kuhn-Tucker, Lagrangean Multiplier, VaR
Fluctuation of the stock price movement often follows the time series pattern, so that has non constant volatility and often followed the long memory effect existing. In this paper we discuss the portfolio optimization based on mean-Value-at-Risk (VaR) by non constant volatility and the long memory effect. Here, mean of the stock return will be estimated using autoregressive fractional integrated moving average (ARFIMA) models, and the volatility we estimated using generalized autoregressive conditional heteroscedastic (GARCH) models. VaR as the risk measure determined based on mean and non constant volatility that. Furthermore, based on mean-VaR, portfolio optimization problem performed using Lagrangian multiplier, and the solution done using Kuhn-Tucker method. The result of the formulation will be used to analyze some stock data that traded in the Indonesian capital market.