Application of GARCH Model to Forecast Data and Volatility of Share Price of Energy (Study on Adaro Energy Tbk, LQ45)
Abstract
Most of the times, Economic and Financial data not only become highly volatile but also show heterogeneous variances (heteroscedasticity). The common method of the Box Jenkins cannot be used for data modeling as the method has an effect of heteroscedasticity (ARCH effects). One of the usable methods to overcome the effect of heteroscedasticity is GARCH model. The aim of this study is to find the best model to estimate the parameters, to predict the share price, and to forecast the volatility of data share price of Adaro Energy Tbk, Indonesia, from January 2014 to December 2016. The study also discuss the Window Dressing. The best model which fits the data is identified as AR(1)-GARCH (1,1). The application of this best model for forecasting the share price of Adaro Energy Tbk, Indonesia, for the next 30 days showed very promising results and the Mean Absolute Percentage Error (MAPE) was determined as 2.16%. Keywords: Volatility, heteroscedasticity, ARCH effect, GARCH model, Window DressingJEL Classifications: C5, Q4, Q47Downloads
Download data is not yet available.
Downloads
Published
2018-05-08
How to Cite
Virginia, E., Ginting, J., & Elfaki, F. A. (2018). Application of GARCH Model to Forecast Data and Volatility of Share Price of Energy (Study on Adaro Energy Tbk, LQ45). International Journal of Energy Economics and Policy, 8(3), 131–140. Retrieved from https://econjournals.com./index.php/ijeep/article/view/6352
Issue
Section
Articles