Error in computing likelihood for initial parameter values

[Juliane]

Hi,

I have a problem with estimation. After loading the data, I get the following error message:

______________________________________________________________
Loading 42 observations from datanew.mat

Error in computing likelihood for initial parameter values
??? Error using ==> print_info at 67
This case shouldn't happen. Contact the authors of Dynare

Error in ==> initial_estimation_checks at 87
print_info(info)

Error in ==> dynare_estimation_1 at 320
initial_estimation_checks(xparam1,gend,data);

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

Error in ==> soe1bayes at 1189
dynare_estimation(var_list_);

Error in ==> dynare at 102
evalin('base',fname) ;
________________________________________________________________
Any idea what it refers to?

Thanks,
Juliane

[pakocica]

Dear Juliane,

try to check if all the parameters are between the lower and upper bound. When you specify the priors for the parameters, you specify mean, std, lb (lower bound),ub (upper bound) and optionally you can specify an initial value. I guess that you specified the initial value for at least one of the parameters outside of the interval [lb,ub]. If this is not the case, please, show us your code.
Best,
Pavel

[MichelJuillard]

If you are calibrating or estimating correlation between shocks, most likely the covariance matrix is not positive definite.
We need to fix the error message in this case.
If you don't have correlated shocks in your model, please, upload the *.mod file and the data so that we can reproduce the problem.

Kind regards

Michel

[evropeec]

I encountered this problem (by accident) when trying to employ the Choleski orthogonalization in a VAR where the variables where generated by a model with a number of shocks smaller than the number of endogenous variables in the VAR. So if you're using simulated data you need to be sure that the number of shocks is sufficiently large relative to the number of endogenous variables. If you're using real data, some of your variables may be highly correlated and need to be either dropped or replaced.

[agupta28]

Dear Michel,

I am simulating data with uncorrelated shocks using the attached file simple3_est.mod
after running this file save the simulated date by filename ``Simulated_data_est3.mat"

I then try and estimate the simulated data using the attached mod file simple2_est.mod

one of the parameters ``lambda" is estimated to be near zero and I get a similar error

Could you please look into this. I am using version 4.1.0

Thanks and Regards,
Abhishek

******************************************************The Error*********************************

??? 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 1050
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 ==> simple2_est at 138
dynare_estimation(var_list_);

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

[edji84]

Hello,

I have a problem with my code for estimating a model to peruvian economy. I don´t understand why mi code don´t work. Please, i really need help.
I ´m using windows 7 (or Ubuntu 12.04) with Dynare 4.2.5
When I run my code, Matlab (or Octave) say:

Error in computing likelihood for initial parameter values
??? Error using ==> print_info at 40
Blanchard Kahn conditions are not satisfied: no stable equilibrium

Error in ==> initial_estimation_checks at 101
print_info(info, options_.noprint)

Error in ==> dynare_estimation_1 at 122
initial_estimation_checks(xparam1,gend,data,data_index,number_of_observations,no_more_missing_observations);

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

Error in ==> Fiscal_ejs_FanNR_est at 302
dynare_estimation(var_list_);

Error in ==> dynare at 120
evalin('base',fname) ;

Eduardo.