cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Try the Materials Informatics Toolkit, which is designed to easily handle SMILES data. This and other helpful add-ins are available in the JMP® Marketplace
Choose Language Hide Translation Bar
kbatch
Level I

Fit Group and Fit Y by X

Hello! I'm attempting to script an analysis where the user inputs a data table, and the script runs a fit Y by X with a user selected column through a column dialog box as Y and all other columns (which can be any mix of continuous and nominal variables) as the X selections. Is there a way to use fit Y by X and Fit Group together to get this all-in-one window and easily allow for the results to be ordered by goodness of fit?

 

This is a script I wrote that outputs them all in one box, but I want the user to easily find the significant fits by ANOVA. Ideally, I would like to output significant fits to a tabular report window, but that is above my knowledge right now. Any help would be appreciated!

Names Default To Here( 1 );
dt = Current Data Table();
dlg = Column Dialog(
	iv = ColList( "Depedent Variable",
		Min Col( 1 ),
		Max Col( 1 ),
		Data Type( "Numeric" )
	)
);
If( dlg["Button"] != 1,
	Beep();
	Throw();
);
iv = dlg["iv"];
n_cols = N Cols( dt );
New Window( "Report",
	H List Box(
		For( c = 1, c <= n_cols, c++,
			Fit Y by X(
				Y( Column( iv[1] ) ),
				X( Column( c ) ),
				Fit Line( 1 ),
				Anova( 1 )
			)
		)
	)
);

 

1 ACCEPTED SOLUTION

Accepted Solutions
txnelson
Super User

Re: Fit Group and Fit Y by X

The script below, runs the Fit Y by X and then output the Summary of Fit table for all of the analyses.  I hope this is what you want.

Names Default To Here( 1 );
dt = Current Data Table();
dlg = Column Dialog(
	iv = ColList( "Depedent Variable",
		Min Col( 1 ),
		Max Col( 1 ),
		Data Type( "Numeric" )//,
		//modeling type(ordinal,nominal)
	)
);
If( dlg["Button"] != 1,
	Beep();
	Throw();
);
iv = Char( dlg["iv"][1] );

// Get the names of all of the columns in the data table
colNameList = dt << get column names( string );
// Remove the Dependent Column from the list
Remove From( colNameList, Contains( colNameList, iv ), 1 );


ft = Fit Y by X(
	Y( Column( iv ) ),
	X( Eval( colNameList ) ),
	Fit Line( 1 ),
	Anova( 1 )
);
(ft << report)[1]["Summary of Fit", Table Box( 1 )] << make combined data table;

Here are a couple of notes:

  1. When you post JSL, please use the JSL icon to open the JSL window and copy the code into that window.  It allows for a cleaner display of the entry.
  2. You need to take the time to read the Scripting Guide, which is available in the JMP Documentation Library, under the Help pull down menu.
  3. The Column Dialog() structure has been depreciated.  You need to move to using a New Window with a Col List Box().
Jim

View solution in original post

2 REPLIES 2
txnelson
Super User

Re: Fit Group and Fit Y by X

The script below, runs the Fit Y by X and then output the Summary of Fit table for all of the analyses.  I hope this is what you want.

Names Default To Here( 1 );
dt = Current Data Table();
dlg = Column Dialog(
	iv = ColList( "Depedent Variable",
		Min Col( 1 ),
		Max Col( 1 ),
		Data Type( "Numeric" )//,
		//modeling type(ordinal,nominal)
	)
);
If( dlg["Button"] != 1,
	Beep();
	Throw();
);
iv = Char( dlg["iv"][1] );

// Get the names of all of the columns in the data table
colNameList = dt << get column names( string );
// Remove the Dependent Column from the list
Remove From( colNameList, Contains( colNameList, iv ), 1 );


ft = Fit Y by X(
	Y( Column( iv ) ),
	X( Eval( colNameList ) ),
	Fit Line( 1 ),
	Anova( 1 )
);
(ft << report)[1]["Summary of Fit", Table Box( 1 )] << make combined data table;

Here are a couple of notes:

  1. When you post JSL, please use the JSL icon to open the JSL window and copy the code into that window.  It allows for a cleaner display of the entry.
  2. You need to take the time to read the Scripting Guide, which is available in the JMP Documentation Library, under the Help pull down menu.
  3. The Column Dialog() structure has been depreciated.  You need to move to using a New Window with a Col List Box().
Jim
kbatch
Level I

Re: Fit Group and Fit Y by X

Thank you very much!