Bootstrap testing for threshold cointegration in threshold vector error correction models

This paper develops a test for the presence of threshold cointegration in a threshold vector error-correction model with the linear no cointegration null hypothesis. We propose a simple algorithm to obtain maximum likelihood estimation of the complete threshold cointegration model. We adopt a SupLM type test; derive its null asymptotic distribution and present bootstrap approximation. Simulation evidence shows that bootstrap inference generates moderate size and power of the test. Our method is illustrated with used U. S. treasury yield curve rates.