The RATS Software Forum

Hello,
I have a question. I am analysing time series data using cointegration and VECM. All the series were tested for a unit root allowing for structural breaks. The tests reveal that all the series are non-stationary, and also contain structural breaks. This suggests that I will need to account for the breaks in the VECM model. However, the structural breaks in all the series occurred at different dates. As a result, I am not sure how to incorporate the different breaks in the VECM. I am asking if you could guide me on the best approach to dealing with structural breaks in a VECM with different break dates.

You seem to be misinterpreting the results of the cointegration test. It's a unit root test allowing for breaks, not a test for the existence of breaks, nor a method of locating breaks. At any rate, you can use Bai, Lumsdaine and Stock to look for common breaks in a multivariate model.

Tom
Thank you so much for the reply. I now understand that a test for unit root allowing for structural breaks is not a test for the presence of breaks. However, aside using the Bai-Perron and the Bai, Lumsdaine, and Stock methods to look for structural breaks, if I test the variables for cointegration with structural breaks using for example, the Gregory-Hansen test, can the breaks in the cointegration test be accounted for in the VECM in the form of dummies?

Not to my knowledge. The Gregory-Hansen test allows for a break in the cointegrating vector. However, they never show any relationship between that and a VECM with any form of break---instead, they simply start with the assumptions about the cointegrating vector.