I have got few questions about VECM and cointegration test.

Basically I conducted a cointegration test on two time series (spot vs forward price) by using the Johansen procedure. The results suggest the presence of a cointegrating relationship between the two prices, with the test statistic significant at 1,5 and 10% level.

In addition I tested for the presence of a 1 to 1 relationship between the series, by testing the restriction on the cointegrating vector Beta = (1,-1).Also in this case I failed to reject the null at all the significance levels.

On this basis, I thought to estimate the corresponding restricted VECM. I based the lag selection on AIC and BIC. However, in both cases diagnostic test on the residuals are quite "catastrophic"suggesting autocorrelation, non-normality and heteroschedasticity.

At this point I was wondering wich are the implications on my analysis and what I should conclude about the cointegrating relationship between the two variables.

Thanks a lot for your help !

  • $\begingroup$ Check out Silvapulle & Podivinsky "The effect of non-normal disturbances and conditional heteroskedasticity on multiple cointegration tests" (2000). $\endgroup$ – Richard Hardy Apr 9 '15 at 14:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.