Exploring the Forecasting Performance of ARIMA-GARCH- Family and Regime Switching ARIMA Models for Industrial Manufacturing in Pakistan
Keywords:Forecasting, ARIMA, GARCH, Asymmetric, TAGRCH, EGARCH, PARCH, MRS- ARIMA
Forecasting plays a vital role in making effective planning and decisions for policy making in almost every field of life. Modeling the dynamic behavior of price series due to non-stationarity, conditional heteroscedasticity, leverage effect and structural breaks is challenging. This opens the doors to the applications of non-linear models such as Markov Regime Switching, Symmetric and asymmetric generalized autoregressive conditional heteroscedastic (GARCH) models along with commonly used Autoregressive Integrated Moving Average (ARIMA) models. The main aim of this study is to explore the estimating and forecasting performance of Regime Switching ARIMA(MRS-ARIMA) models and ARIMA models with symmetric GARCH, and asymmetric GARCH (EGARCH, TGARCH and PARCH) models for the annual industrial manufacturing output prices in Pakistan. The empirical evidence based on the application of these models to the selected price series revealed that the Markov regime switching model successfully captures the heteroscedasticity depicting the powerfulness of these models. The forecasting performance of asymmetric GARCH models is better than the symmetric GARCH model. Within the family of GARCH models, the ARMA (2, 1)-PARCH (1, 1) perform the best. Overall, MRR- ARMA models provide the best predictive ability among all the models based on AIC. The use of regime switching models should be increased due to the ability to capture structural changes, heteroscedasticity and non-linearity simultaneously.
How to Cite
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.