paper examines the structural changes emanating from Gross domestic product of
Nigeria from 1980 to 2017. Out of sample forecast performances of non-linear
time series SETAR model were examined. All necessary theoretical frameworks
were stated and stationarity tests conducted before the model setting.
Out-of-sample forecast performances between the standard linear ARIMA model and
non-linear SETAR model were compared. The Empirical illustration shows that the
non-linear SETAR model has superior forecasting power than linear ARIMA model
using Gross domestic product of Nigeria. It suffices to recommend the
non-linear model for would be policy makers, investors and academia for
forecasting. However, this does not foreclose the fact that the linear ARIMA forecast
model could still be used by forecasters in the absence of SETAR and other
powerful non-linear model.
Gross domestic product of Nigeria, SETAR Model, in
and out samples forecast, ARIMA model