Publications Details

Publications / Conference Poster

Parmest: Parameter Estimation Via Pyomo

Klise, Katherine A.; Nicholson, Bethany L.; Staid, Andrea; Woodruff, David L.

The ability to estimate a range of plausible parameter values, based on experimental data, is a critical aspect in process model validation and design optimization. In this paper, a Python software package is described that allows for model-based parameter estimation along with characterization of the uncertainty associated with the estimates. The software, called parmest, is available within the Pyomo open-source software project as a third-party contribution. The software includes options to obtain confidence regions that are based on single or multi-variate distributions, compute likelihood ratios, use bootstrap resampling in estimation, and make use of parallel processing capabilities.