ARIMA forecast of Nigerian inflation rates with Covid-19 pandemic event in focus

Abstract

We modelled Nigeria Inflation rate using univariate time series methodology. A data set covering January 2003 to June 2022 was divided into two time periods, vis-à-vis, Pre COVID-19 pandemic event period and the Combined period (combination of both pre-COVID-19 and COVID-19 periods). This was done with a view to examining the effect of COVID-19 pandemic on the data generation mechanism of Nigeria Inflation rate. The Inflation rate data was found to be non-stationary and was appropriately differenced to attain stationarity. ARIMA models were fitted for the two scenarios using Box Jenkins methodology. Appropriate diagnostic analysis was conducted and the estimated models were used for forecasting. While the estimated model for Pre COVID-19 event series yielded forecast rates that suggest an onward rise in Nigeria Inflation rate, the Combined period model yielded forecast rates that suggest a downward movement of the Inflation rate.

Keywords

• inflation rate
• arima
• covid-19 pandemic
• box-jenkins methodology
• forecasting