January 2019
·
168 Reads
·
1 Citation
Studies in Computational Intelligence
In the last two decades, Nigeria’s economy has been affected by several shocks such as the 1985-86 oil price crash; 1997 Asian financial crisis; 2008-2009 global financial crisis, oil price crash that started in 2014 as well as political uncertainties in the country. Parameters of econometric models are dependent on prevailing policy and will react to policy changes. Yet, previous researches on trade-growth modelling in Nigeria had assumed parameter constancy over time. Thus, this paper constructed a time varying parameter model for trade-growth nexus and demonstrated how the model can be useful in the detection of structural breaks and outliers. The paper first demonstrated via the rolling regression method that the parameters have been time dependent and proceeded with the Kalman filter to estimate the transition of the changing parameters of the trade-growth nexus. Thereafter, it presented applications to show how the auxiliary residuals of the model can be used to detect the time of structural breaks and outliers.