Macroeconomic variables play important role in shaping the economy of every nation. Most times researchers assume that there are interaction among them but not in all cases. This work is an attempt to study if there exist any interaction among prices of copper, maize and oil. Data on these selected variables ranging from April 1971 to July 2024 were carefully extracted from central bank of Nigeria’s website (www.cbn.org.ng). Unlike other macroeconomic variables that exhibits interdependence among themselves, this study has demonstrated price independence among copper, maize and oil. Result of the analysis shows stationary series, differencing was done to take care of stochastic disturbance term. Cointegration test was deemed not necessary. The work adopts lag 1 as suggested from the lag selection result output. Table 3 displays VAR estimates result that suggests weak interaction among the prices of copper, maize and oil. Their R-squared and adjusted R-squared show a very weak explanatory power among the three variables. For copper it is 13% and 12% respectively, for maize 7% and 6% respectively and for oil 4% and 3% respectively. These are convincing evidence to say that there is no interaction among the prices of the variables. Table 4 shows VAR residual serial correlation. From the result output, all the probability values are greater than 0.05 at lag 1 so we do not reject Ho. We conclude that there is no serial correlation. Table 5 show VAR residual heteroskedasticity tests, from the result output we observe that the joint chi-square value of 253.2185 with 162 df and P=0.0000 shows significant evidence of homoscedastic. That is, there is no heteroskedasticity. Table 6 shows VAR residual Normality tests, the jarque-Bera test value of 6896.977 is far way greater than the P-values (0.0000) indicating that we do not reject Ho meaning that VAR residuals are normally distributed. Table 7 shows dynamic stability test of the VAR estimates, from the plot it is clearly seen that no root lies outside the unit circle. Hence, VAR satisfies the stability condition. Table 8 shows granger causality/block exogeneity Wald tests. From the result output, there is clearly no variable that granger causes each other because their various chi-square values are greater than 0.05. Copper, maize and oil do not granger cause each other hence there will be no need for impulse response function (IRF) test. In conclusion, as important as these macroeconomic variables are, the researcher recommends that policy makers are to focus more on other variables such as interest rate, unemployment rate, crime rate, inflation, GDP etc to get macroeconomic variables with high interdependence for the purpose of policy formulation for the country.
Macroeconomic, Variables, VAR, Stationary, Multivariate, Time Series
IRE Journals:
Vincent Nchedo Chukwukelo "Vector Autoregressive Modeling of Selected Macroeconomic Variables in Nigeria" Iconic Research And Engineering Journals Volume 9 Issue 4 2025 Page 1279-1286 https://doi.org/10.64388/IREV9I4-1711589-8400
IEEE:
Vincent Nchedo Chukwukelo
"Vector Autoregressive Modeling of Selected Macroeconomic Variables in Nigeria" Iconic Research And Engineering Journals, 9(4) https://doi.org/10.64388/IREV9I4-1711589-8400