Title: Unit root testing for the Buffered autoregressive model
Authors: Wai-Keung Li - The University of Hong Kong (Hong Kong) [presenting]
Di Wang - University of Hong Kong (Hong Kong)
Abstract: Buffered (hysteretic) auto-regression is an extension of the classical threshold autoregressive model by allowing a buffered region for regime changes. We study asymptotic statistical inference for the 2-regime buffered autoregressive (BAR) model with possible unit roots. A sup-LR test is proposed for testing for the nonlinear buffer effect in the possible presence of unit roots, and a class of unit root tests is proposed to identify the number of nonstationary regimes under the BAR model. The wild bootstrap is suggested to approximate the critical values of the tests. Two real examples are considered to illustrate the proposed methodology.