Restrictions on coefficients for estimate() function

Hi all,

I would like to impose coefficient constraints for following state space model (contains 3 NaN's)

Mdl = ssm(A,B,C,D);

with using Observation deflators 'Predictors' and 'Beta0' in following equation:

params0 = [0.45 0.45 0.2]; %initial values for the three NaN's [EstMdl,estParams] = estimate(Mdl,y,params0,'Predictors',Z,... 'Beta0',...)

1) According the ssm, the first two NaN's of Matrix A which are estimated with estimate() should be identical with the first two coefficients of 'Beta'. How should i write this?

2) Some of the 'Predictor' coefficients should be zero. How can I implement this?

3) Is it possible to constrain certain coefficients to sum up to 1, when i cannot give them identifying names but only NaN's?

Any help would be much appreciated. Happy to share the underlying code.

Thank you Oli