st: Estimation VARMA model in state space form--not cancave

[xuyongdeng <xuyongdeng@hotmail.com>]

I am estimating a three variables VARMA(1,1) model. I firstly transfor the
VARMA model into a state space model, where it seems to be a 6 state
equations and 3 observed equations model. I try several times, but each time
no matter how many iteration it is done, the log likelyhood function always
appears not concave.
Does anyone know what the problem is?

The programme are as follow:
space (u1 l.u1 l.u2 l.u3 l.u4 l.u5 l.u6 e.u1,state)
 >      (u2 l.u1 l.u2 l.u3 l.u4 l.u5 l.u6 e.u2,state)
 >      (u3 l.u1 l.u2 l.u3 l.u4 l.u5 l.u6 e.u3,state)
 >      (u4 e.u1,state noconstant)
 >      (u5 e.u2,state noconstant)
 >      (u6 e.u3,state noconstant)
 >      (duraion u1 u2 u3, noconstant) 
>       (volume u1 u2 u3, noconstant) 
>       (returnsqr u1 u2 u3, noconstant)
>       constraint(1/6)  covstate(diagonal) 

duration volume and returnsqr are three observed variables
-- 
View this message in context: http://statalist.1588530.n2.nabble.com/Estimation-VARMA-model-in-state-space-form-not-cancave-tp5469640p5469640.html
Sent from the Statalist mailing list archive at Nabble.com.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/