Example of Profiling a Fitted Loglinear Variance Model
This example demonstrates the use of the Prediction Profiler to find factor settings that achieve a specific target for the response while minimizing variance. Fit the models and then use the Profiler to match a target value for a mean and to minimize variance.
1. Select Help > Sample Data Folder and open InjectionMolding.jmp.
2. Select Analyze > Fit Model.
Since the variables in the data table have been assigned preselected roles, the analysis runs automatically.
3. Click the Loglinear Variance Fit red triangle and select Profilers > Profiler.
4. Click the Prediction Profiler red triangle and select Optimization and Desirability > Set Desirabilities.
5. In the Response Goal window that appears, change Maximize to Match Target.
6. Click OK.
7. In the second Response Goal window, click OK.
8. Click the Prediction Profiler red triangle and select Optimization and Desirability > Maximize Desirability.
9. Click the Prediction Profiler red triangle and select Prediction Intervals.
Note: Your results might vary due to random starting values in the optimization process.
Figure 11.7 Profiler to Match Target and Minimize Variance with Prediction Intervals
One of the best ways to see the relationship between the mean and the variance is by looking at the individual prediction intervals. Confidence intervals for a mean response (like those shown by default in the Prediction Profiler) do not show information about the variance model as well as individual prediction confidence intervals do. The prediction intervals show that when predicting a new response, the variability of the response, Shrinkage, increases as Hold Time increases.