Dynare forums

Dear all,

I have a download file from professor frank schorfheide's website (mod and data files are attached below). When I tested it, it reports the error messages as follows. I have no idea why. When I run the non-demeaned model, it works perfect. Could any of you help me spot the problem?

Second, I want to know if I want put 'endogenous_prior' option in estimation block. Do I need to specify prior density in 'estimated_params' block as usual? It does not seems right to me.


Many thanks!

Best,

Richard





POSTERIOR KERNEL OPTIMIZATION PROBLEM!
(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!
> In dynare_estimation_1 at 694
In dynare_estimation at 89
In us_m1_demeaned at 190
In dynare at 180





Error using chol
Matrix must be positive definite.

Error in metropolis_hastings_initialization (line 68)
d = chol(vv);

Error in random_walk_metropolis_hastings (line 62)
[ ix2, ilogpo2, ModelName, MetropolisFolder, fblck, fline, npar, nblck, nruns,
NewFile, MAX_nruns, d ] = ...

Error in dynare_estimation_1 (line 782)
feval(options_.posterior_sampling_method,objective_function,options_.proposal_distribution,xparam1,invhess,bounds,dataset_,options_,M_,estim_params_,bayestopt_,oo_);

Error in dynare_estimation (line 89)
dynare_estimation_1(var_list,dname);

Error in us_m1_demeaned (line 190)
dynare_estimation(var_list_);

Error in dynare (line 180)
evalin('base',fname) ;

Try a different mode-finder.

Regarding the endogenous prior: the Christiano Trabandt Walentin endogenous prior is based on "Bayesian learning". It updates a given prior using the first two sample moments. This means that you need to specify an initial prior for updating.

Thanks! I will try it. For endogenous prior option one, appreciate your illustration. It comes much clearer to me now.

Best,

Richard