MODELLING NIGERIA INFLATION RATE USING
AUTOREGRESSIVE INTEGRATED MOVING AVERAGE
(ARIMA) MODEL
Author(s) : Isiaka Zainab Olanshile, Nasiru Mukaila Olakorede
ABSTRACT:
This study is about modelling and forecasting inflation rates in Nigeria using time Tseries ARIMA model. The study fit a univariate time series ARIMA model to the Nigeria inflation rate from 1999 to 2022. The analysis reveals a decreasing trend in inflation from 1999 to 2000, followed by a peak from 2005 to 2006 and a consolidation period from 2008 to 2017. Afterward, there is a gradual upward trend that continues to increase at different stages over time until 2022. Autoregressive Integrated Moving Average (ARIMA) model was estimated and the best model is ARIMA (0,0,1), Akaike Information Criteria (AIC) of 125.61 and R2 of 0.02329. The model was further validated with L-jung Box test and the residuals of the fitted ARIMA (0,0,1) model are white noise and uncorrelated, indicating that the model accurately fits the series. The study also provides forecasts for three years from 2023 to 2025, which indicate a steady increase in inflation rates in Nigeria.
KEYWORD(S):
Inflation Rate, ARIMA, Forecast, Stationarity, Non-Stationarity, White
Noise