- Cover Page
- Title Page
- Copyright Page
- Contents
- Learn About JMP
- Introduction to Designing Experiments
- Examples Using the Custom Designer
- Building Custom Designs
- Creating a Custom Design
- Creating Random Block Designs
- Creating Split Plot Designs
- Creating Split-Split Plot Designs
- Creating Strip Plot Designs
- Special Custom Design Commands
- Save Responses and Save Factors
- Load Responses and Load Factors
- Save Constraints and Load Constraints
- Set Random Seed: Setting the Number Generator
- Simulate Responses
- Save X Matrix: Viewing the Number of Rows in the Moments Matrix and the Design Matrix (X) in the Log
- Optimality Criterion
- Number of Starts: Changing the Number of Random Starts
- Sphere Radius: Constraining a Design to a Hypersphere
- Disallowed Combinations: Accounting for Factor Level Restrictions
- Advanced Options for the Custom Designer
- Assigning Column Properties
- How Custom Designs Work: Behind the Scenes
- Screening Designs
- Response Surface Designs
- Full Factorial Designs
- Mixture Designs
- Discrete Choice Designs
- Space-Filling Designs
- Accelerated Life Test Designs
- Nonlinear Designs
- Taguchi Designs
- Evaluating Experimental Designs
- Augmented Designs
- Prospective Sample Size and Power
- References
- Index
Chapter 12
Nonlinear Designs
Design of experiments with models that are nonlinear in their parameters is available using either the DOE menu or the JMP Starter DOE category.
Nonlinear designs offer both advantages and disadvantages compared to designs for linear models.
On the positive side, predictions using a well chosen model are likely to be good over a wider range of factor settings. It is also possible to model response surfaces with more curvature and with asymptotic behavior.
On the negative side, the researcher needs a greater understanding of both the system and of the nonlinear design tool.
-
No Comment
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.