摘要：This paper investigates the finite-sample properties of the smooth transi-tion-based cointegration test proposed by Kapetanios et al. (2006) when the datagenerating process under the alternative hypothesis is a globally stationary secondorder LSTR model. The provided procedure describes an application to long-runequilibrium relations involving real exchange rates with symmetric behaviour. Weutilise the properties of the double LSTR transition function that features unit rootbehaviour within the inner regime and symmetric behaviour in the outer regimes.Hence, under the null hypothesis we imply no cointegration and globally station-ary D-LSTR cointegration under the alternative. As a result of the identificationproblem the limiting distribution derived under the null hypothesis is non-standard.The Double LSTR is capable of producing three-regime TAR nonlinearity when thetransition parameter tends to infinity as well as generating exponential-type non-linearity that closely ap proximates ESTR nonlinearity. Therefore, we find that theDouble LSTR error correction model has power against both of these alternatives
关键词：Cointegration tests; LSTR; Monte carlo simulation; Nonlinear error;correction
