Threshold variable selection via a L 1 penalty approach

Abstract

Selecting the threshold variable is a key step in building a general threshold autoregressive (TAR) model. Based on a general smooth threshold autoregressive (STAR) model, we propose to select the threshold variable by the recently developed L 1-penalizing approach. Moreover, by penalizing the direction of the coefficient vector instead of the coefficients themselves, the threshold variable is more accurately selected. Oracle properties of the estimator are obtained. Its advantage is shown with both numerical and real data analysis.