topic Help with Kernel Smoother output in Discussions

Help with Kernel Smoother output

gutloja — Fri, 19 Feb 2016 01:30:37 GMT

[Kernel smoother - output in column]

Hello All,

I am running a kernel smoother but I would like to have the output of each smoother in a column, using the name of the variable. I still would like to see the Fit Group graph, for visual inspection, but I can't find a way to have JMP write the predicted values to a different column automatically [I usually save predicteds, coefficients, etc. manually].

Is there a way to do this?. This is the script I am using:

KernelSmootherGroup# = Expr(
Fit Group(
    Bivariate( Y( :Name( "Var1" ) ), X( :RowNO ),
        Kernel Smoother( 1, 1, 0.0985804416403786, 0 )
    ),
    Bivariate( Y( :Name( "Var2" ) ), X( :RowNO ), 
        Kernel Smoother( 1, 1, 0.0985804416403786, 0 )
    ),
    Bivariate( Y( :Name( "Var3" ) ), X( :RowNO ), 
        Kernel Smoother( 1, 1, 0.0985804416403786, 0 )
    ),
    <<{Arrange in Rows( 1 )},
    SendToReport(
        Dispatch(
            {"Bivariate Fit of Var1 By RowNO"},
            "2",
            ScaleBox,
            {Min( 0.1 ), Max( 1 ), Inc( 0.02 ), Minor Ticks( 1 )}
        ),
        Dispatch(
            {"Bivariate Fit of Var1 By RowNO"},
            "Bivar Plot",
            FrameBox,
            {Frame Size( 1085, 240 )}
        ),
        Dispatch(
            {"Bivariate Fit of Var2 By RowNO"},
            "2",
            ScaleBox,
            {Min( 0.1 ), Max( 1 ), Inc( 0.02 ), Minor Ticks( 1 )}
        ),
        Dispatch(
            {"Bivariate Fit of Var2 By RowNO"},
            "Bivar Plot",
            FrameBox,
            {Frame Size( 1092, 240 )}
        ),
        Dispatch(
            {"Bivariate Fit of Var3 By RowNO"},
            "2",
            ScaleBox,
            {Min( 0.1 ), Max( 1 ), Inc( 0.02 ), Minor Ticks( 1 )}
        ),
        Dispatch(
            {"Bivariate Fit of Var3 By RowNO"},
            "Bivar Plot",
            FrameBox,
            {Frame Size( 1096, 240 )}
        ),
    )
);    
    
);
KernelSmootherGroup#;

Any help will be appreciated,

JoseGL

Re: Help with Kernel Smoother output

txnelson — Fri, 19 Feb 2016 13:08:05 GMT

Here is your script, modified with the code added that will output the Predicted values:

Names default to here(1);

dt=current data table();

KernelSmootherGroup# = Expr(

Fit Group(

biv1=Bivariate( Y(:Name( "Var1" ) ), X( :RowNO ), Kernel Smoother(1, 1, 0.0985804416403786, 0 ) ),

biv2=Bivariate( Y(:Name( "Var2" ) ), X( :RowNO ), Kernel Smoother(1, 1, 0.0985804416403786, 0 ) ),

biv3=Bivariate( Y(:Name( "Var3" ) ), X( :RowNO ), Kernel Smoother(1, 1, 0.0985804416403786, 0 ) ),

<<{Arrange in Rows( 1 )},

SendToReport(

Dispatch(

{"Bivariate Fit of Var1 By RowNO"},

"2",

ScaleBox,

{Min( 0.1 ), Max( 1 ), Inc( 0.02 ), Minor Ticks( 1 )}

Dispatch(
{"Bivariate Fit of Var1 By RowNO"}, "Bivar Plot", FrameBox, {Frame Size( 1085, 240 )} ),

Dispatch(

{"Bivariate Fit of Var2 By RowNO"},

"2",

ScaleBox,

{Min( 0.1 ), Max( 1 ), Inc( 0.02 ), Minor Ticks( 1 )}

Dispatch(
{"Bivariate Fit of Var2 By RowNO"}, "Bivar Plot", FrameBox, {Frame Size( 1092, 240 )} ),

Dispatch(

{"Bivariate Fit of Var3 By RowNO"},

"2",

ScaleBox,

{Min( 0.1 ), Max( 1 ), Inc( 0.02 ), Minor Ticks( 1 )}

Dispatch(
{"Bivariate Fit of Var3 By RowNO"}, "Bivar Plot", FrameBox, {Frame Size( 1096, 240 )} ),

)

);

biv1 << (curve[1] << save predicteds);
biv2 << (curve[1] << save predicteds);
biv3 << (curve[1] << save predicteds);

);

KernelSmootherGroup#;

Re: Help with Kernel Smoother output

gutloja — Fri, 19 Feb 2016 17:51:43 GMT

Thanks a lot, Jim.
This is what I was looking for.