Re: st: cointegration - value of the speed of adjustment

[Prakash Singh <prakashbhu@gmail.com>]

Dear Sabina
The error term coefficient indeed represents speed of adjustment and
the value of coefficient tells us the percent of correction happening
in first to going by this meaning the coefficient should not be
smaller that -1.
Now, coming to your second query Johansen is extension of
Engle-Granger procedure for multivariate sense so the requirement has
to be same.


Hope this helps.

Regards
Prakash
On Sun, Jan 13, 2013 at 8:22 PM, Sabina Kummer <sabina.noo@postmail.ch> wrote:
> Dear all,
>
> I am currently working with error correction models and I have a question regarding the value of the error correction term’s coefficient. I analyze cointegration between two variables according to the Engle-Granger approach and to the Johansen approach.
>
> The coefficient of the error correction term represents the speed of adjustment to the long-term equilibrium. I know that it must be negative and significantly different from zero (otherwise the error correction model can not be validated). My question is the following: must the error correction term’s coefficient be higher than -1 (the value must be between -1 and 0)? What if it is lower than -1? Does that mean that the adjustment is too strong and thus, that the long term equilibrium will never be reached?
> Do both procedures (Engle-Granger and Johansen) have the same requirements regarding the value of this coefficient?
>
> Thank you very much for your help and your consideration.
>
> Kind regards,
>
> Sabina
>
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/