Simulation experiments on the performance of structural change tests in cointegration

Author: Khedhiri Sami  

Publisher: Taylor & Francis Ltd

ISSN: 0094-9655

Source: Journal of Statistical Computation and Simulation, Vol.78, Iss.7, 2008-01, pp. : 585-593

Disclaimer: Any content in publications that violate the sovereignty, the constitution or regulations of the PRC is not accepted or approved by CNPIEC.

Previous Menu Next

Abstract

We present Monte Carlo simulation experiments to investigate the finite-sample properties of structural change tests in cointegrated relationships. These tests are computed using three alternative efficient estimators of cointegrating vectors. It is shown that the tests have overall accurate size and good power performance with a slight out performance of the fully modified and the leads and lags estimators-based tests. However our simulation results reveal substantial power decrease for all the tests due to higher error serial correlation.