Modeling and Forecasting Inflation in Zimbabwe: a Generalized Autoregressive Conditionally Heteroskedastic (GARCH) approach

Of uttermost importance is the fact that forecasting macroeconomic variables provides a clear picture of what the state of the economy will be in future (Sultana et al, 2013). Nothing is more important to the conduct of monetary policy than understanding and predicting inflation (Kohn, 2005). Inflation is the scourge of the modern economy and is feared by central bankers globally and forces the execution of unpopular monetary policies. Inflation usually makes some people unfairly rich and impoverishes others and therefore it is an economic pathology that stands in the way of any sustainable economic growth and development. Models that make use of GARCH, as highlighted by Ruzgar & Kale (2007); vary from predicting the spread of toxic gases in the atmosphere to simulating neural activity but Financial Econometrics remains the leading discipline and apparently dominates the research on GARCH. The main objective of this study is to model monthly inflation rate volatility in Zimbabwe over the period July 2009 to July 2018. Our diagnostic tests indicate that our sample has the characteristics of financial time series and therefore, we can employ a GARCH – type model to model and forecast conditional volatility. The results of the study indicate that the estimated model, the AR (1) – GARCH (1, 1) model; is indeed an AR (1) – IGARCH (1, 1) process and is not only appropriate but also the best. Since the study provides evidence of volatility persistence for Zimbabwe’s monthly inflation data; monetary authorities ought to take into cognisance the IGARCH behavioral phenomenon of monthly inflation rates in order to design an appropriate monetary policy.