? mydgges

Hi All,

I get the following error message. It happens after computing the mode and before launching the MCMC. For some choice of priors it does not occur. Does anybody know what this might mean?



??? MYDGGES requires two square real matrices of the same dimension.

Error in ==> dr1 at 288
[ss,tt,w,sdim,dr.eigval,info1] = mjdgges(e,d,options_.qz_criterium);

Error in ==> resol at 87
[dr,info] = dr1(dr,check_flag);

Error in ==> dynare_resolve at 34
[oo_.dr,info] = resol(oo_.steady_state,0);

Error in ==> DsgeLikelihood at 117
[T,R,SteadyState,info] = dynare_resolve(bayestopt_.restrict_var_list,…

Error in ==> csminit at 105
f = eval([fcn ‘(dxtest,varargin{:})’]);

Error in ==> csminwel at 153
[f3 x3 fc retcode3] = csminit(fcn,x,f,gcliff,0,eye(nx),varargin{:});

Error in ==> dynare_estimation at 336
[fval,xparam1,grad,hessian_csminwel,itct,fcount,retcodehat] = …

Error in ==> lam at 463

Error in ==> dynare at 81
evalin(‘base’,fname) ;

Most likely there was a problem at the mode computation stage and the values passed to MCMC aren’t a good initial point.
Can you upload the *.mod file and the data so that we can get a closer look?



Thanks Michel! Please find attached the data and *.mod file.


dataaus.m (12.5 KB)
lam.mod (13.9 KB)