Dynare forums

Good afternoon,

I am trying to estimate a model with procyclical leverage in the banking sector. Whether I use mode_compute 4, 6, or 9, I get the following error message:

Error using chol
Matrix must be positive definite.

I have tried to run an identification analysis, in order to see if some parameters are not correctly identified; however, I get:

==== Identification analysis ====

Testing prior mean
Evaluating simulated moment uncertainty ... please wait
Doing 492 replicas of length 300 periods.
Simulated moment uncertainty ... done!
Error using .*
Matrix dimensions must agree.

Error in identification_analysis (line 225)
deltaM = deltaM.*abs(params');

I have searched the forum but I didn't see this issue arising a lot. Please note that my shocks are not correlated. Could someone help?

Some of the variables name you use could be in conflict with matlab standard commands, try to change names.

Sorry, but identification currently does not support measurement errors. The only way to do this currently is to specify the measurement error as a separate structural shock. See the discussion on measurement error specification in Pfeifer(2013): "A Guide to Specifying Observation Equations for the Estimation of DSGE Models" https://sites.google.com/site/pfeiferecon/Pfeifer_2013_Observation_Equations.pdf on how to do this.

Thanks Jpfeifer, this solved my problem.