Hello everybody and merry Christmas!
I know this has been questioned here numerous times, but having done everything I read in this forum the problem is still there. I have transformed the data as the observation equations in log deviations, I have more shocks than observables, I checked the model_diagnostics and also the identification command, and there is no problem. I keep getting the following well known error:
initial_estimation_checks:: The forecast error variance in the multivariate Kalman filter became singular.
initial_estimation_checks:: This is often a sign of stochastic singularity, but can also sometimes happen by chance
initial_estimation_checks:: for a particular combination of parameters and data realizations.
initial_estimation_checks:: If you think the latter is the case, you should try with different initial values for the estimated parameters.
ESTIMATION_CHECKS: There was an error in computing the likelihood for initial parameter values.
ESTIMATION_CHECKS: If this is not a problem with the setting of options (check the error message below),
ESTIMATION_CHECKS: you should try using the calibrated version of the model as starting values. To do
ESTIMATION_CHECKS: this, add an empty estimated_params_init-block with use_calibration option immediately before the estimation
ESTIMATION_CHECKS: command (and after the estimated_params-block so that it does not get overwritten).
That means that your model implies a linear combination between the observables. It often occurs if you observe all components of the budget constraint or of the interest parity condition. Check what happens if you drop one observable at a time.
Thanks a lot for your advice, I tried to drop one observable at a time and I dropped r_mhat variable , but I still keep getting the following well known error:
The forecast error variance in the multivariate Kalman filter became singular.
when I run the estimation command with dynare 4.4.3, start to run estimate command,after it was shown that estimation successful and mode_check plots were shown then it suddenly stopped. I keep getting the following error:
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.
when I check the identification command, I keep getting the following error:
WARNING !!!
The rank of J (moments) is deficient!
Find the reason for stochastic singularity, i.e. the linear combination implied by your model.
Make sure you do only estimate parameters that are identified.
Dynare 4.4.3 did not check for the problem in 1. That is the reason you are now getting the error message.
Thanks a lot for guiding me, after dropping observable variables i could estimate parameters but only use one observable variable and i dropped most of parameters in estimate block to identify parameters.
There is also an important issue about irf. when I stimulated variables with stoch_simul () command , the IRF of the variables fluctuate in a zig-zag manner, I do not know it arises because of some timing error with respect to predetermined variable or the Euler equations or might be a matter of parameters.