Linearized version of baseline dsge model

linear_baseline.mod (10.4 KB) Hi all

I am currently writing up the Fernandez-Villaverde (2006) baseline dsge model in a log-linearized version as a practice. I modified from the non-linear example available in dynare forum. During the modification to linearized version, I encounter the following problems:

  1. in non-linearized version, the steady-state of ld (labor demand) is calculated using fsolve in the “NK_baseline_steadystate.m”. However, in the linearized version, certain equation (e.g. equation 17, market clearing, line 233 of linear_baseline.mod ) depends on the steady state value of ld. Is it possible for me to calculate the steady state value (refer to the variable ld_ss) in the mod file?

  2. Then I just get the steady-state value of ld from the non-linear example and put it directly as in line 178 and errors have occurred. Could anyone please kindly guide me out of the errors… Thank you!!

The steady-state is independent of whether or not you log linearize the model. So the steady-state values of the non-linear model should work in the log-linearized model if the log-linearization is correct.

Sure, you can compute the steady-state using a non-linear model and use those values in your log-linear model. You may find this useful. Four ways to compute the steady state of a DSGE model | Willi Mutschler

It seems there are more problems with the mod file that you need to fix though. Model dynamics shows the following…

>> model_diagnostics(M_,options_,oo_)
MODEL_DIAGNOSTICS: The following endogenous variables aren't present at the current period in the model:
k
warning: Some of the parameters have no value (gammma1, Rbar, Lambdax) when using model_diagnostics. If these parameters are not initialized in a steadystate file or a steady_state_model-block, Dynare may not be able to solve the model...
STEADY:  numerical initial values or parameters incompatible with the following equations
    1    2    3    5    6    7   11   13   15   16   17   20   21   22   27
Check whether your model is truly linear. Put "resid(1);" before "steady;" to see the problematic equations.