Skip to content
Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time


This is a slim Python wrapper to Dynare. It runs the *.mod file, displays the resulting output, and then provides access to the Dynare workspace. It works with both, Matlab and Octave.

For fellow linux users (and probably mac users as well), this also provides a script to execute Dynare from comand line.

NOTE: Since I do not have to use dynare anymore, this is no longer under development! Before, this was at a very early stage and not well-tested (and not tested at all on Windows). Comments via the bug tracker are still welcome.


The declared goal of this project is to provide access to Dynare without having to learn and use Matlab/Octave. This reflects the acknowledgement that the translation of the Dynare codebase to a more efficient open source alternative is a major effort which may take a while. Meanwhile, pynare encourages young researchers to focus on modern open source languages and, as the whole interface is pure Python, still allows them to use Dynare without Matlab.


Package import and running a modfile:

from pynare import pynare
modfile = '/path/to/greatmodel.mod'
mod0 = pynare(modfile)

This will give you the usual info about how the Blanchard-Kahn conditions are messed up. The mod0-object contains all the stuff from dynare. Access the attributes with:

mod0.workspace                   # will give you the matlab workspace
print(mod0.workspace['beta'])    # it contains all the parameters
>>> 0.995                       # likely to be the case

I also added the oo_ Dynare-object, which contains a lot of usefull stuff. Note that you could have also accessed it via mod.workspace['oo_'].

mod0.oo_.keys()              # it is a dict, so browse it via `keys()`
irfs = mod0.oo_['irfs']      # for instance get your impulse response functions (it's again a dict). Maybe plot them?

Just evoking pynare from the comand prompt/shell is as simple as

pynare modelpath/crazy.mod

Installation via pip

Installation is as simply as running (Windows users from the Anaconda Prompt):

pip install pynare

Of course, installing Dynare remains your responibility.

If you care for the bleeding edge developing version (which, at this stage, is recommended), the handy way is to install via the git-command. If you hav not yet installed git, Windows users get it here: (Linux users just use the repo of their distro). Then:

pip install git+


A Python inteface to Dynare with access to its workspace.



No releases published


No packages published