Bootstrap-Based Unit Root Tests for Higher Order Autoregressive Models with GARCH (1, 1) Errors

• Activity Number: 246
• Type: Contributed
• Date/Time: Monday, August 4, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Xiao Zhong*+ and V. A. Samaranayake
• Companies: and Missouri University of Science & Technology
• Keywords: Non-stationary ; Conditional volatility ; Residual bootstrap ; Time series

Abstract

Bootstrap-based unit root tests are a viable alternative to asymptotic distribution-based tests and, in some cases, are preferable because of the serious size distortions the latter tests display under certain situations. While several bootstrap-based unit root tests exist for ARMA processes with homoscedastic errors, only one such test is available when the innovations are conditionally heteroskedastic. The utility of this test is limited because it is restricted to autoregressive processes of order one. We extend this test to autoregressive processes of higher orders and study the finite sample performance of the test using Monte-Carlo simulation. Results show that the proposed tests have reasonable power and size properties.