steady state computations

[uvcindy]

Dear all,
I am using dynare to do the bayesian estimation, but when I use it to estimate some parameters which make a huge difference for the steady state values, the code will give some error, saying, "Impossible to find the steady state, either the model doesn't have a unique steady state or the guess value are too far away from the solution".
I am thinking to create file_steadystate.m to compute the steady state with different parameters, but the question is: can I use fsolve in the .m file and return the steady state values to Dynare code? and how can I feed in the initial guess?
(here not like the example fs2000a_steadystate.m, most of the equations are nonlinear)

thanks

[MichelJuillard]

*_steadystate.m is a standard Matlab function, you can use fsolve inside it to solve a subset of equations that don't have an analytical solution. You just indicate the initial values when you call fsolve

Kind regards

Michel