gEconpy.model.model.Model#

class gEconpy.model.model.Model(variables, shocks, equations, steady_state_relationships, steady_state_equations, ss_solution_dict, param_dict, hyper_param_dict, deterministic_dict, calib_dict, priors, is_linear=False, mode=None, error_func='squared')#

A Dynamic Stochastic General Equilibrium (DSGE) Model.

Stores model primitives (variables, parameters, shocks, equations) and compiles steady-state and linearization functions lazily on first use. Construction accepts raw sympy objects; all graph building and pytensor compilation is deferred to the point of use so that only the graphs actually needed are ever built.

Methods

`Model.__init__`(variables, shocks, equations, ...)	Initialize a DSGE model from sympy primitives.
`Model.equation_tensors`([filter_known])	Pytensor graphs for the model's steady-state equations.
`Model.evaluate_residual`(ss_dict, param_dict)	Evaluate the steady-state residual at given variable and parameter values.
`Model.get`(name)	Get a model variable or parameter by name.
`Model.linearize_model`([order, ...])	Linearize the model around the deterministic steady state.
`Model.param_tensors`([include_free, ...])	Pytensor scalar variables for the model's parameters.
`Model.parameters`(**updates)	Compute the full set of free parameters for the model, including deterministic parameters.
`Model.solve_model`([solver, log_linearize, ...])	Solve for the linear approximation to the policy function via perturbation.
`Model.ss_tensors`([filter_known])	Pytensor scalar variables for the model's steady-state symbols.
`Model.steady_state`([how, use_jac, use_hess, ...])	Solve for the deterministic steady state of the DSGE model.
`Model.symbolic_linearization`([order, ...])	Return the symbolic pytensor graphs for the linearized Jacobian matrices.

Attributes

`backward_variables`	Variables that appear at t-1 in at least one equation (state variables).
`calibrated_params`	List of calibrated parameters in the model, stored as `sympy.Symbol` objects.
`deterministic_params`	List of deterministic parameters in the model, stored as `sympy.Symbol` objects.
`dr_order`	Decision-rule order classifying variables and equations into block-triangular form.
`eq_order`	Row permutation reordering equations as `[static \| lag_only \| lead_only \| both]`.
`equations`	List of equations in the model, stored as Sympy expressions.
`f_params`	Compiled function mapping free parameter values to the full parameter dictionary.
`f_ss`	Compiled function mapping parameters to known steady-state values, or None if no analytical solutions.
`forward_variables`	Forward-looking (jump) variables.
`hyper_params`	List of hyperparameters in the model, stored as `sympy.Symbol` objects.
`inv_eq_order`	Inverse of `eq_order`.
`inv_var_order`	Inverse of `var_order`.
`lead_var_idx`	Column indices of forward-looking variables in the Jacobian matrices.
`n_backward`	Number of backward-looking (state) variables.
`n_forward`	Number of forward-looking (jump) variables.
`n_symbolic_forward`
`n_variables`	Number of endogenous variables in the model.
`param_priors`	Dictionary of prior distributions for the model parameters.
`params`	List of parameters in the model, stored as `sympy.Symbol` objects.
`shock_priors`	Dictionary of prior distributions for the model shocks.
`shocks`	List of shocks in the model, stored as Sympy symbols.
`steady_state_relationships`	List of model equations, evaluated at the deterministic steady state.
`symbolic_forward_variables`	Variables appearing at t+1 in any equation (purely syntactic).
`sympy_to_pytensor_cache`	Cache mapping sympy symbol identifiers to pytensor graph nodes.
`var_order`	Column permutation reordering variables as `[static \| pred_only \| mixed \| forward_only]`.
`variables`	List of variables in the model, stored as Sympy symbols.