Profiler

Example 1

Summary: Simulate a Profiler for the Yield response using the First-Order Kinetics dataset, incorporating normal distributions for Reaction Temperature and Reaction Time factors, and setting desirability functions with specified limits and goals.

Code:

// Open data table
dt = Open("$Sample_Data/First-Order Kinetics.jmp");
// Profiler Simulator
Profiler(
    Y( :Yield ),
    Profiler(
        1,
        Confidence Intervals( 1 ),
        Desirability Functions( 1 ),
        Yield <<
        Response Limits(
            {Lower( 0.5, 0.05 ),
            Middle( 0.6, 0.75 ),
            Upper(
                0.625, 0.815624999999997
            ), Goal( Maximize ),
            Importance( 1 )}
        ),
        Term Value(
            Reaction Temperature(
                525.82376091265
            ),
            Reaction Time(
                0.298,
                Min( 0.101 )
            )
        ),
        Simulator(
            1,
            Factors(
                Reaction Temperature <<
                Random(
                    Normal weighted(
                        525.82376091265,
                        1
                    )
                ),
                Reaction Time <<
                Random(
                    Normal weighted(
                        0.298, 0.03
                    )
                )
            ),
            Responses(
                Yield << No Noise
            )
        )
    ),
    Expand
);

Example 2

Summary: Generate a Profiler report for three response variables (GP Fit, NL Fit, and Difference) with confidence intervals and specific term values using the Nonlinear Examples/CES Production Function.jmp data table.

Code:

// Open data table
dt = Open("$Sample_Data/Nonlinear Examples/CES Production Function.jmp");
// Profiler
Profiler(
    Y( :GP Fit, :NL Fit, :Difference ),
    Profiler(
        1,
        Confidence Intervals( 1 ),
        Term Value(
            l( 0.5 ),
            k( 0.3966 )
        )
    ),
    Expand,
    SendToReport(
        Dispatch( {"Prediction Profiler"},
            "Profiler", FrameBox,
            Frame Size( 174, 87 )
        ),
        Dispatch( {"Prediction Profiler"},
            "Profiler", FrameBox( 2 ),
            Frame Size( 174, 87 )
        ),
        Dispatch( {"Prediction Profiler"},
            "Profiler", FrameBox( 3 ),
            Frame Size( 174, 87 )
        ),
        Dispatch( {"Prediction Profiler"},
            "Profiler", FrameBox( 5 ),
            Frame Size( 174, 87 )
        ),
        Dispatch( {"Prediction Profiler"},
            "Profiler", FrameBox( 6 ),
            Frame Size( 174, 87 )
        ),
        Dispatch( {"Prediction Profiler"},
            "Profiler", FrameBox( 7 ),
            Frame Size( 174, 87 )
        )
    )
);

Example 3

Summary: Set up and configure a Profiler for simulating the Yield response with specified desirability functions and stochastic factors using Normal weighted random distributions for Reaction Temperature and Reaction Time.

Code:

// Open data table
dt = Open("$Sample_Data/Stochastic Optimization.jmp");
// Profiler Ready to Simulate
Profiler(
    Y( :Yield ),
    Profiler(
        1,
        Confidence Intervals( 1 ),
        Desirability Functions( 1 ),
        Yield <<
        Response Limits(
            {Lower( 0.5, 0.05 ),
            Middle( 0.6, 0.75 ),
            Upper( 0.625, 0.815625 ),
            Goal( Maximize ),
            Importance( 1 )}
        ),
        Term Value(
            Reaction Temperature( 530 ),
            Reaction Time(
                0.2,
                Min( 0.101 )
            )
        ),
        Simulator(
            1,
            Factors(
                Reaction Temperature <<
                Random(
                    Normal weighted(
                        530, 1
                    )
                ),
                Reaction Time <<
                Random(
                    Normal weighted(
                        0.2, 0.03
                    )
                )
            ),
            Responses(
                Yield << No Noise
            )
        ),
        Desirability Functions( 0 ),
        Simulator( 0 )
    ),
    Expand
);