TIME SERIES ANALYSIS OF QUARTERLY INFLATIONARY RATE IN NIGERIA

Abstract

 The ultimate goal is to investigate the trend of quarterly inflationary rate in Nigeria. The specific objectives were to estimate model parameters, select the most parsimonious model among candidate models as well as to forecast on the basis of the fitted model. The data used in this study was extracted from www.nigerianstat.gov.ng. The consumer price index for food sector which span from January 2009 to July 2019 was used. A dataset of one hundred and twenty (120) cases was used to build the model while seven (7) cases were used for testing the model. The linear trend, non- seasonal ARIMA (1,2,0) and seasonal ARIMA (1,2,0)(1,0,0)12models were used for forecasting after training the data sets. In adjusting for stationarity, the series was found to be stationary after second finite differencing. Augmented Dickey Fuller (ADF) test was used to assess the stationarity with a p-value of 0.042. The ACF and PACF plot were assessed and it was found that the differenced series followed autoregressive terms. The analysis suggests that ARIMA models can be used to model Nigeria consumer price index. There is a rising trend in the consumer price index inflation of food sector in Nigeria. The seasonal ARIMA (1,2,0)*(1,0,0)12 is more parsimonious than Non-seasonal ARIMA (1,2,0) and linear trend 12 equation. Also, the seasonal ARIMA (1, 2, 0)*(1, 0, 0) is most suitable for forecasting consumer price index as measure of inflation in Nigerian Food Sector. Findings of this study should enable the orientation of policies targeted towards curbing the increasing inflationary rate in the food sector of Nigeria.