Study on Time Series Modeling and Forecasting Inflation: Evidence from Nigeria

A major concern of entrepreneurs and monetary authorities in Nigeria in the past decades was
successful prediction general price level movements. The results allow successful planning on the
part of monetary authorities and continued profit drive on the part of entrepreneurs and investors. This
study uses a univariate model in the form of Autoregressive Integrated Moving Average model
developed by Box and Jenkins and multivariate time series model in the form of Vector
Autoregressive model to forecast inflation for Nigeria. This paper use changes in monthly consumer
price index obtained from the National Bureau of Statistics and the Central bank of Nigeria over the
period 2003 to 2012 to predict movements in the general price level. Based on different diagnostic
and evaluation criteria, the best forecasting model for predicting inflation in Nigeria is identified. The
results will enable policy makers and businesses to track the performance and stability of key
macroeconomic indicators using the forecasted inflation. The study forecasted core inflation using
VAR for the month of 2012:10 to be 11.06 percent. A major limitation of this study is that it focused on
two major forecast tool the VAR method and the ARIMA method and neglected the use of neural
network analysis. In addition, only core inflation was used as a measure of inflation. Subsequent
studies on inflation forecasting in Nigeria should attempt to forecast inflation across a wider spectrum
of inflation measures.