Bayesian Estimation "矩阵必须为正定矩阵。"e

I met this problem when I ran this model using bayesian estimation.
Please help me. Thank you.

错误使用 chol
矩阵必须为正定矩阵。
出错 posterior_sampler_initialization (line 84)
d = chol(vv);
出错 posterior_sampler (line 60)
posterior_sampler_initialization(TargetFun, xparam1, vv,
mh_bounds,dataset_,dataset_info,options_,M_,estim_params_,bayestopt_,oo_);
出错 dynare_estimation_1 (line 465)
posterior_sampler(objective_function,posterior_sampler_options.proposal_distribution,xparam1,posterior_sampler_options,bounds,dataset_,dataset_info,options_,M_,estim_params_,bayestopt_,oo_);
出错 dynare_estimation (line 105)
dynare_estimation_1(var_list,dname);
出错 hjtd0511_bayes.driver (line 1151)
oo_recursive_=dynare_estimation(var_list_);
出错 dynare (line 293)
evalin(‘base’,[fname ‘.driver’]) ;

Without the files it is impossible to tell. Search the forum, there are numerous posts. Most importantly, have a look at the mode_check-plots.

Now, the error are as follows:
Error in computing likelihood for initial parameter values
错误使用 print_info (line 32)
Blanchard & Kahn conditions are not satisfied: no stable equilibrium.
出错 print_info (line 32)
error(message);
出错 initial_estimation_checks (line 185)
print_info(info, DynareOptions.noprint, DynareOptions)
出错 dynare_estimation_1 (line 164)
oo_ = initial_estimation_checks(objective_function,xparam1,dataset_,dataset_info,M_,estim_params_,options_,bayestopt_,bounds,oo_);
出错 dynare_estimation (line 105)
dynare_estimation_1(var_list,dname);
出错 hjtd0511_bayes.driver (line 1171)
oo_recursive_=dynare_estimation(var_list_);
出错 dynare (line 293)
evalin(‘base’,[fname ‘.driver’]) ;

just add

predetermined_variables
R ku;

You model has a unit root, so you are missing diffuse_filter. Also, your observation equations are incorrect. The level of the model variables and their counterparts in the data do not match.

Yes yes. I add. But the problem is same. I don’t find this one that your observation equations are incorrect. And don’t understand this “The level of the model variables and their counterparts in the data do not match.” Please say it specially. Thank you.

For example, in the model, y has steady state 1.6, in the data it is mean 0. Moreover, the data goes to 1e4, which suggests you forgot to take logs.

You mean that steady stata fo y is 0?

Please inllustrate it specially. For example, you can use screenshot. Thank you.

Now your data loading is wrong, I guess. There are only three observations. Does a simulated version of the model work?

It is also that problem.
Error in computing likelihood for initial parameter values
错误使用 print_info (line 32)
Blanchard & Kahn conditions are not satisfied: no stable equilibrium.
出错 print_info (line 32)
error(message);
出错 initial_estimation_checks (line 185)
print_info(info, DynareOptions.noprint, DynareOptions)
出错 dynare_estimation_1 (line 164)
oo_ =
initial_estimation_checks(objective_function,xparam1,dataset_,dataset_info,M_,estim_params_,options_,bayestopt_,bounds,oo_);
出错 dynare_estimation (line 105)
dynare_estimation_1(var_list,dname);
出错 hjtd0511_bayes.driver (line 1171)
oo_recursive_=dynare_estimation(var_list_);
出错 dynare (line 293)
evalin(‘base’,[fname ‘.driver’]) ;

You don’t open my data?

Please try this one.
data.mat (689 Bytes)

You are not listening.

Also:

I add.

You mean that I update y in the data.

Now I am lost. Which problem is still current and what is the most recent set of files?