gEconpy.model.statespace.DSGEStateSpace

class gEconpy.model.statespace.DSGEStateSpace(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)

Core class for estimating DSGE models using PyMC.

Methods

DSGEStateSpace.__init__(variables, shocks, ...)	Create a pmx.statespace.PyMCStateSpace model representing a linearized DSGE.
DSGEStateSpace.build_statespace_graph(data)	Given a parameter vector theta, constructs the full computational graph describing the state space model and the associated log probability of the data.
DSGEStateSpace.configure(observed_states[, ...])	Configure the statespace model for estimation.
DSGEStateSpace.make_symbolic_graph()	Build the symbolic statespace graph for the DSGE model.
DSGEStateSpace.sample_autocorrelation_matrices(idata)	Posterior distribution of the model-implied autocorrelation matrices.
DSGEStateSpace.set_coords()	Provide coordinates to the model.
DSGEStateSpace.set_parameters()	Provides parameter metadata to the model.
DSGEStateSpace.set_shocks()	Provide shock metadata to the model.
DSGEStateSpace.set_states()	Provide state metadata to the model.
DSGEStateSpace.to_pymc([exclude_priors])

Attributes

lead_var_idx	Column indices of forward-looking variables (variables appearing at t+1 in any equation).
n_forward	Number of forward-looking variables.
param_dims	Dictionary of named dimensions for each model parameter