方法对比

	TAR / SETAR：用于机制转换时间序列的门限自回归模型	光滑转换自回归 (STAR) 模型	阈值回归
领域	计量经济学	计量经济学	计量经济学
方法族	Regression model	Regression model	Regression model
起源年份	1990	1994	2000
提出者	Howell Tong	Teräsvirta (1994); van Dijk, Teräsvirta & Franses (2002)	Bruce E. Hansen
类型	Nonlinear time-series model with regime switching	Nonlinear time-series regime-switching model	Nonlinear regime-switching regression
开创性文献	Tong, H. (1990). Non-linear Time Series: A Dynamical System Approach. Oxford University Press. ISBN: 978-0-19-852300-6	Teräsvirta, T. (1994). Specification, Estimation, and Evaluation of Smooth Transition Autoregressive Models. Journal of the American Statistical Association, 89(425), 208–218.	Hansen, B. E. (2000). Sample Splitting and Threshold Estimation. Econometrica, 68(3), 575-603.
别名	Threshold Autoregression, Self-Exciting Threshold Autoregression, SETAR Model, Eşik Otoregresyon	smooth transition autoregressive model, LSTAR, ESTAR, logistic STAR	threshold model, regime-switching regression, sample splitting model, Eşik Değer Regresyonu (Threshold Regression)
相关	2	4	5
摘要	TAR and SETAR are nonlinear autoregressive models introduced by Howell Tong (1990) that allow a time series to follow different linear dynamics in distinct regimes, separated by one or more threshold values. SETAR is the self-exciting variant, in which the threshold variable is a lagged value of the series itself, making it particularly suited to cycles, asymmetric adjustment, and limit-cycle behavior observed in economic and financial data.	The Smooth Transition Autoregressive (STAR) model is a nonlinear time-series model, developed in Teräsvirta's 1994 framework, that lets the dynamics move smoothly rather than abruptly between two regimes. The logistic variant (LSTAR) captures asymmetric business cycles and the exponential variant (ESTAR) captures purchasing-power-parity deviations.	Threshold regression is a nonlinear, regime-switching model in which the regression parameters take different values above and below an estimated threshold value of a threshold variable. The sample-splitting and threshold-estimation framework was developed by Bruce E. Hansen (2000) and is widely used for time-series and panel data with structural breaks and regime-dependent relationships.