Smooth Transition Autoregressive (STAR) modeling (Teräsvirta, 1994) is an extremely popular approach for nonlinear time series analysis. STAR models, which are a special case of Smooth Transition Regression (STR) models, embed regime-dependent linear auto-regression specifications in a smooth transition nonlinear regression framework.

Similar in concept to discrete Threshold Regression (TR) models, STR models differ in that regime switching occurs smoothly when an observed variable crosses the unobserved thresholds. As a result of this smooth response, STR models are often thought to have more “realistic” dynamics than their discrete TR counterparts. STR models have been applied to a wide range of datasets, from macroeconomic (Teräsvirta and Anderson 1992, Teräsvirta 1994) to the well-known Canadian lynx data (Teräsvirta, 1994).

This section describes EViews tools for estimation of two-regime STR models with unknown parameters for the shape and location of the smooth threshold. EViews estimation supports several different transition functions, provides model selection tools for selecting the best threshold variable from a candidate list, and offers the ability to specify regime varying and non-varying variables and variables that appear in only one regime.

Following estimation, EViews offers specialized views of the transition function and weights and offers tests for linearity against STR alternatives and tests of no remaining nonlinearity and parameter constancy, alongside conventional tests for heteroskedasticity and serial correlation.

Portions of this section offer brief background for STR models. Those desiring additional detail should consult the more detailed discussions in Teräsvirta (1994) and van Dijk, Teräsvirta, and Franses (2002), and the textbook treatment in Martin, Hurn, and Harris (2013, p. 720–726, 744–745).