stability in VECM

[christine]

Dear all,

I'm working on my final paper and i have met some difficulties when estimating a VECM. Using Johansen cointegration test, I've discovered one cointegrating equation between the variables. After that, I estimated the VECM. I've read in some econometrics books that after i estimate the VECM i shall employ a stability test using this path: LAG STRUCTURE- AR ROOTS TABLE. And the roots should be inside the unit circle. However, in my case, the first root is 1 and the others are less than 1. Does this mean that the results I've obtained employing VECM are not good? Or should I have employed the stability test on the VAR from Johansen cointegration test.

Thanks a lot

Christine

[rosen]

hi
does any body have a programme for running recursive residual test or cusum test of a svecm, its urgent.
thanks
rosen

[mirza kaleemullah]

hi Christen

i hope that u have addressed this issue. I am indulged in the same thing that VECM is not stable and every time in estimation same vecm, it gives new results.
anybody there plz?
regards