The issue of stability in VECMs is discussed at length in Johansen's book
on likelihood based inference in cointegrated VARs. ez
****************************************************************
* Eric Zivot phone: 206-543-6715 *
* Department of Economics fax: 206-685-7477 *
* Box 353330 email: [log in to unmask] *
* University of Washington *
* Seattle, WA 98195-3330 *
* *
* www: http://weber.u.washington.edu/~ezivot/homepage.htm *
****************************************************************
On Tue, 31 Aug 1999, Price SG wrote:
> In single equation ECMs the stability condition on the loading is
> obviously pretty simple. But in VECM the condition for
> stability/convergence to the long-run equilibria are more complex. In
> particular, I imagine we might occasionally encounter models where the
> estimated loading(s) in an ECM equation in the VECM has the "wrong" ie
> positive sign (when the level corresponding to the lhs variable is
> normalised with a positive coefficient), but the system is still stable.
>
> Does anyone know any papers that discuss this?
>
>
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|