When you construct a design using the DOE platforms, column properties are saved to the resulting design table. However, some of these column properties are useful in general modeling situations. To use the properties more generally, you can specify them yourself.
Note: Column Properties in the Design of Experiments Guide describes the properties associated with DOE in detail and presents examples. Only a brief description of these properties is provided here.
The following column properties are used in modeling and DOE:
• Response Limits
• Detection Limits
• Design Role
• Coding
• Mixture
• Factor Changes
The Response Limits column property defines a desirability function for the response. The Prediction and Contour Profilers use desirability functions to find optimal settings. See Response Limits in the Design of Experiments Guide.
The Detection Limits column property defines bounds beyond which the response cannot be measured. You can use these limits to specify a censored response in the Generalized Regression platform. See Censoring in Fitting Linear Models.
The Design Role column property indicates how a column is used in both a designed experiment and in the model used to fit the data. For example, a column could represent a continuous factor, a categorical factor, a blocking factor, and so on. See Design Role in the Design of Experiments Guide.
The Coding column property applies a linear transformation to the data in a numeric column. The data are transformed to –1 and +1 based on bounds that you specify. JMP then uses the transformed data values whenever the column is entered as a model effect in the Fit Model platform. See Coding in the Design of Experiments Guide.
The Mixture column property is useful when a column in a data table represents a component of a mixture. The components of a mixture are constrained to sum to a constant. The Mixture column property identifies a column as a mixture component and defines a coding for that column. See Mixture in the Design of Experiments Guide.
The Factor Changes column property indicates how difficult it is to change factor settings in a designed experiment. It is used to create and analyze split-plot, split-split-plot, and two-way split-plot designs. See Factor Changes in the Design of Experiments Guide.