Econometric Approaches to Evaluaing Vector Autoregressive and Autoregressive Distributed Lag Models in Emerging Economies

This study evaluates the performance of Vector Autoregressive (VAR) and Autoregressive Distributed lag (ARDL) Models in analyzing econometric relationships, specifically focusing on the interaction of some macroeconomic indicators in Nigeria. The variables used are inflation, exchange rate, crude oil price, and imports with data from central Bank of Nigeria statistical data base from January, 1995 to March, 2024. Methods utilized are the Vector Autoregressive (VAR) and Autoregressive Distributed Lag (ARDL) model. Output from the analysis reveal that the VAR model demonstrates superior performance, capturing complex, simultaneous lags. It achieves a high explanatory fit of R2- 0.978. In contrast, the ARDL model adopts a more restrictive exogenous framework with simplified lag structure, resulting in lower explanatory power of 21 %. Furthermore, while both models identify a significant relationship between COP and EXR, the VAR model identifies a more pronounced impact (29.11). Findings propose that for the data set analyzed, the VAR framework provides a more robust and comprehensive understanding of the variables’ dynamic interactions. Prioritizing, the VAR over ARDL to describe the relationship between crude oil price and exchange rate is recommended since short-run relationship exist among the variables, policy makers should focus on tactical, immediate interventions rather than relying on long-term structural shifts.