Cointegration Analysis of Multivariate Time Series Model for Inflation Movement in Nigeria

This paper attempt to study the co-integration analysis of multivariate time series model for inflation movement in Nigeria. The data used for this study was extracted from abstract of statistics of Central Bank of Nigeria (CBN) Spanned a period of 10 years (i.e. from 2008-2017). The Johansen co-integration test suggests that there is at least one co-integration vector, which describes the long run relationship between CPI, FPI and NFPI. Furthermore, the accuracy of the model with use of RMSE, MAE, MAPE and Theil’s U statistic shows that the model is accurate for n-step forecasting. Over the time period considered, all the three series showed an increasing pattern, that is, there is sign of non-stationary in each of the series. In order to examine the VAR model, the unit root tests (ADF test), and co-integration analyses were conducted. Unit root tests indicated that all indices are non-stationary at level and are stationary at first difference at 5% significant level. The Johansen co-integration test suggests that there is at least one co-integration vector, which describes the long run relationship between CPI, FPI and NFPI. Furthermore, the accuracy of the model with use of RMSE, MAE, MAPE and Theil’s U statistic show that the model is accurate for n-step forecasting.