Modeling the Error Term by Moving Average and Generalized Autoregressive Conditional Heteroscedasticity Processes: An Advanced Research | Chapter 10 | Innovations in Science and Technology Vol. 4

 The Combine White Noise model beats the existing Generalized Autoregressive Conditional Heteroscedasticity (GARCH) and Moving Average (MA) models in simulating conditional heteroscedasticity and leverage effect errors, according to this study. The MA process is unable to model data that exposes conditional heteroscedasticity, and the GARCH model is likewise unable to describe the leverage effect. The standardised residuals of GARCH errors are decomposed into a series of white noises, which are then modelled using the Combine White Noise model (CWN). With minimum information criteria and high log likelihood values, CWN model estimate produces the best results. When compared to the MA model, the EGARCH model estimate produces better results in terms of minimum information criterion and high log likelihood values. When compared to the GARCH and MA models dynamic evaluation forecast errors, CWN has the smallest forecast errors, indicating the best results. CWN's findings consistently outperform those of GARCH and MA. The CWN provides good results that enhance the flaws of previous models, which is a contribution of this study to the scientific community.

Author(S) Details

Ayodele Abraham Agboluaje
Department of Mathematical Sciences, Faculty of Natural Sciences, Ibrahim Badamasi Babangida University, Lapai, Nigeria.

Suzilah bt Ismail
School of Quantitative Sciences, College of Arts and Sciences, Universiti Utara Malaysia, Malaysia.

Chee Yin Yip
Department of Economics, Faculty of Business and Finance, Universiti Tuanku Abdul Rahman, Malaysia.

View Book:- https://stm.bookpi.org/IST-V4/article/view/5563