Example of Confidence Intervals Using Bagging
Use the bagged standard errors to obtain confidence intervals around predictions for a Neural Network model.
Fit Neural Network
1. Select Help > Sample Data Library and open Tiretread.jmp.
2. Select Analyze > Predictive Modeling > Neural.
3. Select ABRASION and click Y, Response.
4. Select SILICA, SILANE, and SULFUR and click X, Factor.
5. Click OK.
6. (Optional) Enter 123456 next to Random Seed.
Note: Results vary due to the random nature of choosing a validation set in the Neural Network model. To reproduce the results shown in this example, set the Random Seed.
7. Click Go.
Perform Bagging and Launch the Profiler
1. Click the red triangle next to Model NTanH(3) and select Profiler.
2. Click the Prediction Profiler red triangle and select Save Bagged Predictions.
3. Enter 100 next to Number of Bootstrap Samples.
4. (Optional) Enter 123456 next to Random Seed.
Note: Results vary due to the random nature of sampling with replacement. To reproduce the results in this example, set the Random Seed.
5. Click OK.
6. Select Graph > Profiler.
7. Select Pred Formula ABRAISION Bagged Mean and StdError ABRASION Bagged Mean and click Y, Prediction Formula.
8. Select Expand Intermediate Formulas.
9. Click OK.
A JMP Alert window appears asking if you want to use the StdError ABRASION Bagged Mean column to make intervals for the Pred Formula ABRAISION Bagged Mean column.
10. Click Yes in the JMP Alert window.
Figure 3.37 Profiler of Bagged Neural Predictions with Confidence Intervals
A profiler is created with confidence intervals using the values from the StdError ABRASION Bagged Mean column.