gEconpy.model.statespace.DSGEStateSpace.__init__

DSGEStateSpace.__init__(variables, shocks, equations, param_dict, hyper_param_dict, param_priors, shock_priors, parameter_mapping, steady_state_mapping, linearized_system, var_order=None, log_linearized_variables=None, sympytensor_cache=None, filter_type='standard', verbose=True)

Create a pmx.statespace.PyMCStateSpace model representing a linearized DSGE.

Users should not create this class direction, and should instead use gEconpy.model.build.statespace_from_gcn() to compile a statespace model from a gcn file.

Parameters:

variables: list of TimeAwareSymbol

List of variables in the model

shocks: list of TimeAwareSymbol

List of shocks in the model

equations: list of sympy.Expr

List of equations in the model

param_dict: dict

Dictionary of default parameter values, as defined in the model file

hyper_param_dict: dict

Dictionary of default hyperparameter values, as defined in the model file

param_priors: dict

Dictionary of preliz parameter priors

shock_priors: dict

Dictionary of preliz shock priors

parameter_mapping: dict

Symbolic function mapping input parameters to the full vector of parameters, including deterministic.

steady_state_mapping: dict

Symbolic function mapping input parameters to the steady state values of the model

linearized_system: list of pt.TensorVariable

List of four symbolic expressions representing the linearized system of equations as partial jacobians of the model equations with respect to variables at time t+1 (A), t (B), t-1 (C), and with respect to exogenous shocks (D), each evaluated at the (symbolic) steady state.

log_linearized_variables: list of str, optional

Base names of variables that were log-linearized when building linearized_system. Used by configure(ss_obs_intercept=...) to decide whether an observation intercept entry is log(v_ss(p)) (log-linearized) or v_ss(p) (level-linearized).

verbose: bool

If True, show diagnostic messages.