search

Pyomo / pyomo

An object-oriented algebraic modeling language in Python for structured optimization problems.
https://www.pyomo.org
Other
1.92k stars 499 forks source link

Using parmest to estimate experiment specific parameters #2466

Open adowling2 opened 2 years ago

adowling2 commented 2 years ago

Summary

Are there examples of using parmest to simultaneously estimate both "global" parameters across all experiments (stage 1 in the stochastic programming abstraction, i.e., what is done by default when specifying theta) and individual/local parameters for each experiment (stage 2 in the stochastic programming abstraction)?

Rationale

This question is motivated by a reaction kinetics project in which we want to estimate a "global" set of kinetic parameters while also fitting initial conditions for each experiment (with modest bounds).

Description

Were are able to do this by passing the stage 1 parameter values to parmest as theta and leaving the stage 2 parameters as free variables (with bounds). While this works, it does not properly compute the covariance matrix.

Additional information

Links to any examples are greatly appreciated. Otherwise, I'd like to discuss here what would be involved in adding this as a feature.

adowling2 commented 2 years ago

@kaklise @blnicho @DLWoodruff @AlexanderPerez23 @kanishka-ghosh @lxhowl