Daftar Isi: MEMBANDINGKAN PENDEKATAN PARAMETER MAXIMUM LIKELIHOOD ESTIMATION (MLE) DAN MODIFIED PROFILE LIKELIHOOD (MPL) UNTUK MENDAPATKAN MODEL VOLATILITAS EGARCH OPTIMUM

Daftar Isi:

The purpose of this study is to discuss the best approach for EGARCH volatility models by first comparing the two approaches will be tested parameters, namely the Maximum Likelihood Estimation (MLE) and the Modified Profile Likelihood (MPL). Of this study will be obtained some information about the best approach to establish parameters optimum EGARCH models, so as to produce an optimum EGARCH models that have small error value. The type of this research is a quantitative, that used data collection method is a documentation, research data is Composite Stock Price Index in the city of Malang in the span of 1 January 2007 to 31 December 2011,the data taken in Composite Stock Price Index of Machung Universities. To promote the goals of this research, there are some tests that used the Modified Box-Pierce (Ljung Box) and Kolmogrove-Smirnov normality test. The Results of this study declare that parameter Maximum Profile Likelihood (MPL) is better than the parameters Maximum Likelihood Estimation (MLE) for EGARCH model, the EGARCH models are Y_t=-0,277793Y_(t-1)-0,994936Y_(t-2)+ε_t with its residual variance equation as follows ln⁡(σ_(t-1)^2 )=-0,467285-0,290782|e_(t-1) |+0,488910(e_(t-1) )+1,054246ln⁡(σ_(t-1)^2) , where Y_t= is forecasting future stock price, and σ_(t-1)^2= variance of residual