1

I want to create a table that predicts the value of several Y variables given that two X variables have fixed value. I want to do this on a per group basis using Fit Model of JMP. The problem is that per group, I get one table. So if I have 20 groups, I get 20 tables and these tables doesn't even have names for me to identify which group they belong to. How do I output just one table for all? Below is my code:

New Window(
"Fit Model",
V List Box(
    Fit Model(
        Y(
            :Name( "Y var1" ),
            :Name( "Y var2" ),
            :Name( "Y var3" ),
            :Name( "Y var4" )
        ),
        Effects( :Name( "X var1" ), :Name( "X var2" ) ),
        Personality( Standard Least Squares ),
        Emphasis( Effect Leverage ),
        By (:Grpvar1, :Grpvar2), 
        Run(
            Profiler(
                1,
                Confidence Intervals( 1 ),
                Term Value(
                    Name( "X var1" )(63, Lock( 1 ), Show( 1 )),
                    Name( "X var2" )(33, Lock( 1 ), Show( 1 ))
                           ), 
                Output Grid Table (1),
                    )
           )
        ) 

     )
   )
1201ProgramAlarm
  • 32,384
  • 7
  • 42
  • 56
schrodingerscat
  • 189
  • 1
  • 3
  • 15
  • I was thinking of two work-arounds for this. The first is to create a summary table, subset per group, run fit model, and append results to a master table. Unelegant and complicated but I think I can code that one. The second is finding a way to identify which table belongs to which group and putting an extra column for that. Then concatenating all tables. I think this is simpler, but I don't know how to do it yet. – schrodingerscat Feb 16 '17 at 23:34

0 Answers0