![]() |
Dakota Reference Manual
Version 6.15
Explore and Predict with Confidence
|
Use a polynomial surrogate as the discrepancy model
Alias: none
Argument(s): none
Child Keywords:
Required/Optional | Description of Group | Dakota Keyword | Dakota Keyword Description | |
---|---|---|---|---|
Optional | basis_order | Basis function order of the model discrepancy |
For the purposes of model discrepancy, constant, linear, and quadratic polynomial surrogate models are available. These models are less expensive to calculate than Gaussian process models but are consequently less flexible to discrepancy trends that are not constant, linear, or quadratic in nature. For more information, see the Dakota User's and Theory Manuals.
By specifying this command, each discrepany function will take the form of a polynomial regression. They will all have the same order.