ScholarGate
Get started
Regression modelEconometricsNonlinear regressionModel

Panel Smooth Transition Regression

Also known as: Smooth-transition panel model

OriginatorGonzalez, Terasvirta, and van DijkYear2005Sources2Related methods4

Panel Smooth Transition Regression (PSTR) models nonlinear panel relationships where coefficients transition smoothly (rather than abruptly) between regimes as a transition variable crosses thresholds. Introduced by Gonzalez et al. (2005), it extends univariate smooth-transition autoregression (STAR) models to panels, capturing gradual shifts in economic behavior. This approach is realistic when adjustment costs cause smooth (not sudden) regime changes.

Key highlights

  • Captures smooth, realistic regime transitions
  • Endogenously estimates transition speed and threshold location
  • Panel structure improves efficiency relative to individual-unit estimation
  • Economically interpretable regime-specific effects

Intuition

This section is available to Pro members. Upgrade to Pro

How it works

This section is available to Pro members. Upgrade to Pro

When to use it

Use PSTR when you suspect smooth (gradual) regime changes in panel relationships. Examples: how growth and inflation respond to monetary policy depends smoothly on current inflation level, how firm investment responds to credit constraints depends smoothly on leverage ratio. Avoid when regimes change abruptly.

Strengths & limitations

Strengths
  • Captures smooth, realistic regime transitions
  • Endogenously estimates transition speed and threshold location
  • Panel structure improves efficiency relative to individual-unit estimation
  • Economically interpretable regime-specific effects
Limitations
  • Estimation is computationally demanding (nonlinear optimization in high dimensions)
  • Specification of transition variable is critical; sensitivity to this choice
  • Interpretation of regime-specific coefficients when transition is not sharp
  • Standard asymptotics can be unreliable with slow transitions

Common pitfalls

This section is available to Pro members. Upgrade to Pro

Applications

This section is available to Pro members. Upgrade to Pro

Frequently asked

How do I choose the transition variable?

Select based on economic logic. If unsure, test multiple candidates and compare fit via LR tests or information criteria. Plot coefficients against candidate variables to visualize nonlinearity.

What if I have multiple threshold candidates?

Nest models: test linear (no transition), single-transition, and multi-transition specifications. Each test uses likelihood-ratio statistics; choose most parsimonious model that fits well.

How interpret regime-specific coefficients when transition is smooth?

Coefficients apply to extreme values of the transition variable. For intermediate values, effects are weighted averages of regime-specific effects. Simulate effects at various transition-variable values.

Does PSTR require stationarity?

PSTR assumes mean-reversion (stationarity) within each regime. Test stationarity per regime; if integrated, differences or cointegration methods may be needed.

Sources

  1. 1.
    Gonzalez, A., Terasvirta, T., & van Dijk, D. (2005). Panel smooth transition regression models. Research Paper, Melbourne Institute of Applied Economic and Social Research.
    https://econpapers.repec.org/paper/mlseconwp/
  2. 2.
    Terasvirta, T. (1994). Specification, estimation, and evaluation of smooth transition autoregressive models. Journal of the American Statistical Association, 89(425), 208-218.
    DOI 10.1080/01621459.1994.10476462

You have read it. What now?

Ask about this methodSimilar methods

Cite this page

ScholarGate. (2026, June 3). Panel Smooth Transition Regression. ScholarGate. https://scholargate.app/econometrics/panel-smooth-transition-regression