The Profiler (or Prediction Profiler) shows prediction traces for each X variable.
Illustration of Prediction Traces illustrates part of the profiler for the Reactor.jmp sample data table. The vertical dotted line for each X variable shows its current value or current setting. Use the Profiler to change one variable at a time in order to examine the effect on the predicted response.
Illustration of Prediction Traces
The factors F and Ct in Illustration of Prediction Traces are continuous. If the factor is nominal, the x-axis displays the levels.
The horizontal dotted line shows the current predicted value of each Y variable for the current values of the X variables.
A line segment is plotted for each level of the row effect. Response values predicted by the model are joined by line segments. Non-parallel line segments give visual evidence of possible interactions. However, the p-value for such a suggested interaction should be checked before concluding that it exists. Interaction Plots gives an interaction plot matrix for the Reactor.jmp sample data table.
1.
Open the Reactor.jmp sample data table.
2.
Select Analyze > Fit Model.
3.
Select Y and click Y.
4.
Make sure that the Degree box has a 2 in it.
5.
Select Ct, A, T, and Cn and click Macros > Factorial to Degree.
6.
Click Run.
7.
From the red triangle menu next to Response Y, select Factor Profiling > Interaction Plots.
Interaction Plots
The plot corresponding to the T*Cn interaction is the third plot in the bottom row of plots or equivalently, the third plot in the last column of plots. Either plot shows that the effect of Cn on Y is fairly constant at the low level of T, whether Cn is set at a high or low level. However, at the high level of T, the effect of Cn on Y differs based on its level. Cn at –1 leads to a higher predicted Y than Cn at 1. Note that this interaction is significant with a p-value < 0.0001.
Contour Profiler shows a contour profiler view for the Tiretread.jmp sample data table. Run the data table script RSM for 4 Responses and select Profilers > Contour Profiler from the Least Squares Fit report menu.
Contour Profiler
Note: This option appears only if you specify the Macros > Mixture Response Surface option for an effect. For complete details see the Mixture Profiler chapter of the Profilers book.
Mixture Profiler shows the Mixture Profiler for the model in the Plasticizer.jmp sample data table. Run the Model data table script and then select Factor Profiling > Mixture Profiler from the report’s red triangle menu. You modify plot axes for the factors by selecting different radio buttons at the top left of the plot. The Lo and Hi Limit columns at the upper right of the plot let you enter constraints for both the factors and the response.
Mixture Profiler
The Cube Plots option displays predicted values for the extremes of the factor ranges. These values appear on the vertices of cubes (Cube Plots). The vertices are defined by the smallest and largest observed values of the factor. When you have multiple responses, the multiple responses are shown stacked at each vertex.
1.
Open the Reactor.jmp sample data table.
2.
Select Analyze > Fit Model.
3.
Select Y and click Y.
4.
Make sure that the Degree box has a 2 in it.
5.
Select Ct, A, T, and Cn and click Macros > Factorial to Degree.
6.
Click Run.
7.
Cube Plots
Note that there is one cube for Cn = –1 and one for Cn =1. To change the layout so that the factors are mapped to different cube coordinates, click a factor name in the first cube. Drag it to cover the factor name for the desired axis. For example, in Cube Plots, if you click T and drag it over Ct, then T and Ct (and their corresponding coordinates) exchange places. To see the levels of Cn in a single cube, exchange it with another factor in the first cube by dragging it over that factor.
The Box Cox Y Transformation option fits transformations from λ = –2 to 2 in increments of 0.2. To choose a proper value of λ, the likelihood function for each of these transformations is computed. They are computed under the assumption that the errors are independent and normal with mean zero and variance σ2. The value of λ that maximizes the likelihood is selected. This value also minimizes the SSE over the selected values of λ.
Shows the best value for λ but enables you to enter a value of λ of your choice. Selecting this option also creates a column in the data table with the formula for your specified transformation.
1.
Open the Reactor.jmp sample data table.
2.
Select Analyze > Fit Model.
3.
Select Y and click Y.
4.
Make sure that the Degree box has a 2 in it.
5.
Select F, Ct, A, T, and Cn and click Macros > Factorial to Degree.
6.
Click Run.
7.
From the red triangle menu next to Response Y, select Factor Profiling > Box Cox Y Transformation.
The plot in Box Cox Y Transformation shows that the best values of λ are approximately between 1.0 and 1.5. The value that JMP selects, using its 0.2 unit grid of λ values, is 1.2.
8.
To see the 1.2 value, select Save Best Transformation and look at the formula or select Save Specific Transformation and look at the value presented.
Box Cox Y Transformation
Surface Plot shows the Surface Profiler for the model in the Odor.jmp sample data table. Run the Model data table script and then select Factor Profiling > Surface Profiler from the report’s red triangle menu. You can change the variables on the axes using the radio buttons under Independent Variables. Also, you can plot points by clicking Actual under Appearance.
Surface Plot