topic Re: How to save predicted values into matrix or list without saving into existing data table in Discussions

How to save predicted values into matrix or list without saving into existing data table

wu — Wed, 17 Oct 2018 23:13:57 GMT

Names Default To Here( 1 );
dt = Open( "$SAMPLE_DATA/Big Class.jmp" );
obj = Bivariate( Y( :Weight ), X( :Height ) )<<Group by(:sex);;
obj<<Fit spline(0.1, Standardized,{Save Predicteds});

tried lis below , it won't work.

val_predict_ls=obj<<Save Predicteds.

Re: How to save predicted values into matrix or list without saving into existing data table

txnelson — Thu, 18 Oct 2018 02:32:15 GMT

I am not sure how to create the matrix without creating the column in the data table, but here is a method that functionally will do what you want

Names Default To Here( 1 );

dt = Open( "$SAMPLE_DATA/Big Class.jmp" );

// Get the groupings for the sex column
Summarize( dt, bysex = by( :sex ) );
obj = Bivariate( Y( :Weight ), X( :Height ) ) << Group by( :sex );
obj << Fit spline( 0.1, Standardized, {Save Predicteds} );

// Create the matricies for the groups and delete the
// columns from the data table
For( i = N Items( bysex ), i >= 1, i--,
	Eval(
		Parse(
			bysex[i] ||
			"_Predicted_Matrix = column(dt,N Col(dt))<<get values;"
		)
	);
	dt << delete columns( N Cols( dt ) );
);
Show( f_Predicted_Matrix, m_Predicted_Matrix );

Re: How to save predicted values into matrix or list without saving into existing data table

wu — Thu, 18 Oct 2018 17:34:13 GMT

Thanks.
My attempt was trying get the a few key summary values like mean, sigma etc from a group of predicted values based on a large data base.
Looks like, need to go your way by saving into data table, then deleted them.