search

escheffel / pymaclab

Dynamic Stochastic Equilibrium Models (DSGE) in Python
www.pymaclab.com
Other
137 stars 47 forks source link

Improve updater functions and consider adding memoization to mkjahen method #2

Closed escheffel closed 12 years ago

escheffel commented 12 years ago

The DSGE model instances still don't have decent enough update functions which can be called when the model has been changed in some fundamental way, that is either the parameter space or the nonlinear FOCs. Also, when updating the FOCs does happend at runtime, it may be wise to have some sophisticated memoization decorator attached to the methods computing the analytical Jacobian and Hessian.

escheffel commented 12 years ago

The updater functions are now in a usable state, with the proviso that additional wrapped options may have to be included in the future. Memoization of the methods generating the derivatives of the models are not needed at the moment as updater analogues are already available which bypass the analytical differentiation step