Skip to Main Content
Skip Nav Destination
Nonlinear Regression Modeling for Engineering Applications: Modeling, Model Validation, and Enabling Design of Experiments
By
R. Russell Rhinehart
R. Russell Rhinehart
Search for other works by this author on:
ISBN:
9781118597965
No. of Pages:
400
Publisher:
ASME-Wiley
Publication date:
2016

There are four main expectations of a model that claim to properly represent some physical system.

  1. There should not be any trend in residuals w.r.t. any variable (input, output, chronological order, or treatment). The model should go through the center of the data, throughout the entire range.

  2. The variability expectation from propagation of uncertainty in the data model should match the magnitude of the residuals.

  3. The model should pass logical tests. Both asymptotic limits and local trends projected from the model should be consistent with phenomenological expectations. This is for both coefficient and input variable values.

  4. Coefficient values should match expectations from literature – homolog interpolation or extrapolation, similar conditions, and so on.

16.1
Introduction
16.2
Logic-Based Validation Criteria
16.3
Data-Based Validation Criteria and Statistical Tests
16.4
Model Discrimination
16.5
Procedure Summary
16.6
Alternate Validation Approaches
16.7
Takeaway
Exercises
This content is only available via PDF.
You do not currently have access to this chapter.
Close Modal

or Create an Account

Close Modal
Close Modal