Publication date: 07/08/2024

Factor Profiling

In the Fit Least Squares report, the Factor Profiling options (accessed from the Response red triangle menu) help you explore and visualize your estimated model. You can explore the shape of the response surface, find optimum settings, simulate response data based on your specified noise assumptions, and transform the response if needed.

The Profiler, Contour Profiler, Mixture Profiler, and Surface Profiler are extremely versatile tools whose use extends beyond standard least squares models. For more information about their interpretation and use, see Profilers.

If the response column in your model contains an invertible transformation of one variable, the profilers and Interaction Plots show the response on the untransformed scale.

The following Factor Profiling options are available:

Note: If your model contains an expression or vector as an effect, most of these options are not available.

Profiler

Shows prediction traces for each X variable. Enables you to find optimum settings for one or more responses and to explore response distributions using simulation. See Profiler and Profiler in Profilers.

Interaction Plots

Shows a matrix of interaction plots, when there are interaction effects in the model. See Interaction Plots.

Contour Profiler

Provides an interactive contour profiler, which is useful for optimizing response surfaces graphically. See Contour Profiler and Contour Profiler in Profilers.

Mixture Profiler

Shows response contours of mixture experiment models on a ternary plot. See Mixture Profiler and Mixture Profiler in Profilers.

Note: This option appears only if you apply the Mixture Effect attribute to three or more effects or the Mixture property to three or more columns.

Cube Plots

Shows predicted values for the extremes of the factor ranges laid out on the vertices of cubes. See Cube Plots.

Box Cox Y Transformation

Finds a Box-Cox power transformation of the response that is best in terms of satisfying the normality and homogeneity of variance assumptions. See Box-Cox Y Transformation.

Surface Profiler

Shows a three-dimensional surface plot of the response surface. See Surface Profiler and Surface Plot in Profilers.

Want more information? Have questions? Get answers in the JMP User Community (community.jmp.com).