Dsge modelling using dynare

Please have a look at this file. Thanks

RR_LCR_Indo_steadystate.m (8.4 KB)

That file is hard to port to newer Dynare versions due to the excessive use of global variables. Can you stick to Dynare 4.4.3?

Well, do you mean lowering my dynare version to 4.4.3? Could I maybe install this version to work in tandem with the 5.5 version in the same computer? Or would it be possible for I manually convert the 4.4.3 files into the 5.5 version?

You can easily run different Dynare version by selecting the Matlab path to the version you want to run. You are facing the incompatibility introduced in 4.6 documented at

Changing the interface is straightforward, but removing the global parameters is a mess.

By global parameters, are you referring to lines 5 to 26 on the steady state files, or this part

global beta_p beta_e phi hab_e alppha delta…
eps_y eps_l m_e eps_be eps_d delta_b kappa_kb…
hab_p kappa_i kappa_p kappa_w iota_p iota_w …
kappa_be kappa_d rho_a rho_qk rho_k…
c_p l_p w_p c_e s_e r_k k J_r R_b R_be…
K_b B_e D prof_bank B Y t_p tau_p pie pie_wp xx v…
R R_d weight_l q_k invest Y_pol lambda_p lambda_e…
a_e eps_k eps_qk liq_shock tau div_cb g mp_shock…
util_p welfare_p util_e welfare_e tot_welf…
rho_r rho_v xi_pie xi_y xi_v rho_ror rho_mp…
psi_mp psi_rr psi_ror psi_v xi_ror…
RV rho_rr xi_rr omega_rrr omega_rx1 omega_rx2 mu_rr RF_b RF_cb …
RF_t ror R_rr omega_rf R_rf R_x1 R_x2…
Cap_ratio RV_ratio LCR RF_ratio Liq_asset_ratio TA Loan_asset_ratio…
Y_tot X_lcr X_rr c_y i_y g_y B_e_y RF_y RV_y D_y adj_firm…
adj_bank gamma penalty…
pie_ss R_ss B_ss Y_pol_ss mu_rr_ss ror_ss v_ss X_lcr_temp X_rr_temp…
mod_index patient rho_l Rb_markup Rd_markdown mc R_real R_b_real R_d_real…
nom_loan_gr nom_dep_gr nom_bank_cap_gr nom_rv_gr nom_rf_gr rr_shock v_shock ror_shock …
res Liq_asset depo asset Lev_ratio dist_mean dist_sigma RVRF_ratio …
rho_sigma psi_sigma dist_sigma_ss sigma_shock…
rho_mean psi_mean dist_mean_ss mean_shock…

And what you mean by removing these?

You need to check the full codes. Usually, the header and footer of the steady state file will read out the parameter values and pass them back. But there may be a reason why the authors made those parameters global. I don’t know whether that is the case.

Wow, it does not get any better. Thank you very much for your follow up this long. I’ll explore the possibility of having the 4.4.6 version, who know it may be easier after all.
Best greetings!

You should be able to use 4.5.7.

When I search for dynare 4.5.7, I only get dynare 6.1 for download. would you recommend, what is your opinion?

All Dynare versions are at
https://www.dynare.org/release/

Thank you.

I’ve now donwloaded the 4.5.7

then I did call it as addpath c:\dynare\4.5.7\matlab
but got this error:

"addpath c:\dynare\4.5.7\matlab
Warning: Name is nonexistent or not a directory: c:\dynare\4.5.7\matlab

In path (line 109)
In addpath>doPathAddition (line 116)
In addpath (line 80)"

should I have restarted the computer?

Or put it in a special folder?

Dear Jpfeifer,

can you help me look at this error, once I downloaded dynare 4.5.7 and set the appropriate pat I still get this error:
"

addpath c:\dynare\4.5.7\matlab

Warning: Name is nonexistent or not a directory: c:\dynare\4.5.7\matlab

In path (line 109)

In addpath>doPathAddition (line 116)

In addpath (line 80) "

Thank you

Check whether that really is the folder where you installed Dynare 4.5.7 to.

How the initial values are calculated? Can we use the first value of the serie as initial values?

No, the model concepts and the data values will differ because they are in different units. Apart from that, see

1 Like

The article was useful. Thank you.

However, How the initial values are calculated?

If you cannot just guess them based on economic logic, you need to calculate them based on the static equilibrium conditions that define the steady state.

1 Like

That means calculate them manually, with pen and paper. Is that right?

Yes, exactly.

1 Like