chol matrix

[picchio79]

Hello everybody,

I am trying to estimate a simple RBC model for rhe euro area.

The strange thing is that sometimes it works well. Sometimes without modifying anything I get the following:

??? Error using ==> chol
Matrix must be positive definite.

Error in ==> metropolis_hastings_initialization at 52
d = chol(vv);

Error in ==> random_walk_metropolis_hastings at 43
[ ix2, ilogpo2, ModelName, MhDirectoryName, fblck, fline,
npar, nblck, nruns, NewFile, MAX_nruns, d ] = ...

Error in ==> dynare_estimation_1 at 1061
feval(options_.posterior_sampling_method,'DsgeLikelihood',options_.proposal_distribution,xparam1,invhess,bounds,gend,data,...

Error in ==> dynare_estimation at 62
dynare_estimation_1(var_list,varargin{:});

Error in ==> vero at 156
dynare_estimation(var_list_);

Error in ==> dynare at 132
evalin('base',fname) ;


I attach my files.
Hope you can help me to fix it
I thank you in advance ,

Alessandro

[jpfeifer]

Hi, this is a common problem. You will find several posts regarding this issue in the forum. The short answer is that csminwel involves some randomness and sometimes the inverse hessian resulting from the optimization is not positive definite, even when you keep the model, the data and the starting values the same. If you keep getting the error message and mode_compute=4 (the default) does not succeed, use mode_compute=6 and see the Dynare wiki (http://www.dynare.org/DynareWiki/MonteCarloOptimization)