Abstract

In our previous studies, we modified the Enders and Siklos test for threshold error correction to a version allowing the individual threshold variable to be responsible for the asymmetric mechanism of the system. The idea was to learn about the threshold mechanism both in the long and short run. In this paper, we tested for the asymmetry of the adjustment of the error correction mechanism towards the long-run path. The subsamples within regimes differ in size with respect to the threshold value. The novelty lies in the division of both short and long-run variables according to a threshold variable with a given threshold value (assumed or estimated). We named the test extended Enders and Siklos test (exE-S). The present study focuses on the power and size of the modified procedure. A simulation study was designed, computed and conducted. The results are favourable for the proposed approach, although they strongly depend on the difference in values between the adjustment parameters in the regimes.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call