Empirical Modelling and Model Selection for Forecasting Monthly Inflation of Ghana

Abstract

Inflation is the persistent increase in the level of consumer prices or a persistent decline in the purchasing power of money. Inflation is of global concerns because it can distort economic patterns and can result in the redistribution of wealth when not anticipated, thus there is a need to know the pattern of inflation in the country. In this study, we employed an empirical modeling and model selection for monthly inflation in Ghana from January 2009 to December 2013 using the Box-Jenkins approach. The results showed that ARIMA (1, 2, 1) model was appropriate for modelling the inflation rates with a maximum log likelihood value of -64.21, and least AIC value of 134,43, AICc value of 134.87 and BIC value of 140. 61. An ARCH-LM test and Ljung-Box test on the residuals of the models revealed that the residuals are free from heteroscedasticity and serial correlation respectively. Ghana is likely to experience a persistence increase in inflation rate with double digit hence the government should reconsider his monetary policies.

Keywords: Inflation, Box-Jenkins, Empirical, Ghana.