DynareJulia / Dynare.jl

A Julia rewrite of Dynare: solving, simulating and estimating DSGE models.
https://www.dynare.org
Other
95 stars 20 forks source link

Dynare does not precompile on linux(ubuntu) with julia version 1.6.7 #62

Closed ruben-poghosyan closed 9 months ago

ruben-poghosyan commented 9 months ago

ERROR: LoadError: please run Pkg.build("Pardiso") before loading the package Stacktrace: [1] error(s::String) @ Base ./error.jl:33 [2] top-level scope @ ~/.julia/packages/Pardiso/rrIt7/src/Pardiso.jl:6 [3] include @ ./Base.jl:384 [inlined] [4] include_package_for_output(pkg::Base.PkgId, input::String, depot_path::Vector{String}, dl_load_path::Vector{String}, load_path::Vector{String}, concrete_deps::Vector{Pair{Base.PkgId, UInt64}}, source::String) @ Base ./loading.jl:1235 [5] top-level scope @ none:1 [6] eval @ ./boot.jl:360 [inlined] [7] eval(x::Expr) @ Base.MainInclude ./client.jl:446 [8] top-level scope @ none:1

MichelJuillard commented 9 months ago

From the error message, can you try to first install Pardiso and build it then install Dynare? Also, consider installing Julia v1.10, Dynare will run much faster.

ruben-poghosyan commented 9 months ago

From the error message, can you try to first install Pardiso and build it then install Dynare? Also, consider installing Julia v1.10, Dynare will run much faster.

Thank you for the quick reply, I have tried using Dynare with Julia v1.10 but apparently some functions did not work properly(for example the same mod file works on windows but does not work on linux with the exact same environment configuration), that's why I tried using the version specified in the docs, otherwise Dynare installs on Julia v1.10,

ruben-poghosyan commented 9 months ago

here's the model that I am trying to run, but unfortunately it does not produce a result , I would be thankful if you can point out if somethings not supported here `var

L_GDP, GDP_GAP, L_GDP_BAR, G_Y_BAR

;

varexo

E_GDP_GAP, E_L_GDP_BAR, E_G_Y_BAR

;

parameters ALPHA1, ALPHA2, G; ALPHA1 = 0.7; ALPHA2 = 0.9; G = 2.3;

model;

L_GDP = GDP_GAP + L_GDP_BAR;

GDP_GAP = ALPHA1*GDP_GAP(-1) + E_GDP_GAP;

L_GDP_BAR = L_GDP_BAR(-1) + G_Y_BAR/4 - G/4 + E_L_GDP_BAR;

G_Y_BAR = ALPHA2G_Y_BAR(-1) + (1-ALPHA2)G + E_G_Y_BAR;

end;

shocks;

var E_GDP_GAP; stderr 1.8; var E_L_GDP_BAR; stderr 1.2; var E_G_Y_BAR; stderr 0.2;

end;

steady_state_model;

L_GDP = 1; L_GDP_BAR = 1; GDP_GAP = 0; G_Y_BAR = G;

end;

// steady(nocheck);

stoch_simul(order=1, irf=80, nograph);

varobs L_GDP;

calib_smoother(datafile='gdp.csv', diffuse_filter);`

MichelJuillard commented 9 months ago

Purely backward models aren't yet supported in DynareJulia. For linear models, as it is the case here, this will be added shortly.

MichelJuillard commented 9 months ago

From the error message, can you try to first install Pardiso and build it then install Dynare? Also, consider installing Julia v1.10, Dynare will run much faster.

Thank you for the quick reply, I have tried using Dynare with Julia v1.10 but apparently some functions did not work properly(for example the same mod file works on windows but does not work on linux with the exact same environment configuration), that's why I tried using the version specified in the docs, otherwise Dynare installs on Julia v1.10,

You must mean the same mod file works in Dynare Matlab but not in Dynare Julia rather than Windows and Linux

ruben-poghosyan commented 9 months ago

From the error message, can you try to first install Pardiso and build it then install Dynare? Also, consider installing Julia v1.10, Dynare will run much faster.

Thank you for the quick reply, I have tried using Dynare with Julia v1.10 but apparently some functions did not work properly(for example the same mod file works on windows but does not work on linux with the exact same environment configuration), that's why I tried using the version specified in the docs, otherwise Dynare installs on Julia v1.10,

You must mean the same mod file works in Dynare Matlab but not in Dynare Julia rather than Windows and Linux

Yes, thank you for your swift response again

MichelJuillard commented 9 months ago

Purely backward linear models are now supported in DynareJulia with https://github.com/DynareJulia/LinearRationalExpectations.jl/commit/9f1c5cabbbff1b1d19e01b7472c7991e86ef0177

Please update your installation with

using Pkg
Pkg.update()