We compare linear autoregressive (AR) models and self-exciting threshold autoregressive (SETAR) models in terms of their point forecast performance, and their ability to characterize the uncertainty surrounding those forecasts, i.e. interval or density forecasts. A two-regime SETAR process is used as the data-generating process in an extensive set of Monte Carlo simulations, and we consider the discriminatory power of recently developed methods of forecast evaluation for different degrees of non-linearity. We find that the interval and density evaluation methods are unlikely to show the linear model to be deficient on samples of the size typical for macroeconomic data.

Additional Metadata
Keywords forecasting, non-linearity, uncertainty
Persistent URL dx.doi.org/10.1002/for.863, hdl.handle.net/1765/11145
Journal Journal of Forecasting
Clements, M.P, Franses, Ph.H.B.F, Smith, J, & van Dijk, D.J.C. (2003). On SETAR non-linearity and forecasting. Journal of Forecasting, 22(5), 359–375. doi:10.1002/for.863