(minus)hessian matirx and variance decomposition problem

Hello! I was replicating the paper of Peter Ireland (2004), ‘Technology shocks in the new Keynesian model’.
I have several problems when running the following mod file to do ML estimation.

  1. After running the results there is some thing wrong like this.

(minus) the hessian matrix at the “mode” is not positive definite!
=> posterior variance of the estimated parameters are not positive.
You should try to change the initial values of the parameters using
the estimated_params_init block, or use another optimization routine.
Warning: The results below are most likely wrong! [/quote]

I have tried different mode_compute 3,4,8,9. And model identification shows that all parameters are identified.But the problem was still there.

2.When I do sto_simul after estimation, there was a warning like this:

[quote]Warning: Some of the parameters have no value (sig_a, sig_e, sig_z, sig_r)
when using stoch_simul. If these parameters are not initialized in a
steadystate file, Dynare may not be able to solve the model… [/quote]

so how to do simulation in this case?

  1. The shock_decomposition cannot be carried out also. I wonder what i did wrong.

gpr_data.m (7.7 KB)
full.mod (2.52 KB)

Please see github.com/JohannesPfeifer/DSGE_mod/tree/master/Ireland_2004
and the disclaimer at the beginning of the mod-file there.

thank you so much Dr.Pfeifer, it really helps me!
so much appreciation!