How to Pass Variables to the X Parameter in Fit Life by X?

[#1]

Hi everyone,

I need to dynamically pass variables to the X parameter in the fit life by x function. The first part of my code generates the analysis report as expected, but the second part only brings up the startup window without a report.

Here’s what my code looks like:

Names Default To Here(1);

dt = Open( "$sample_data/big class.jmp" );

x_var = "sex";
y_var = "weight";

obj1 = dt << Fit Life by X(
	Distribution( Weibull ),
	Nested Model Tests( Location and Scale ),
	Y( Eval(y_var) ),
	X( :sex ),
);

wait(2);

obj2 = dt << Fit Life by X(
	Distribution( Weibull ),
	Nested Model Tests( Location and Scale ),
	Y( :weight ),
	X( Eval(x_var) ),
);

Has anyone encountered this issue or knows how to correctly pass the variables to the X parameter?

Thanks!

[#2]

This anomaly appears to be a bug. The Nested Model Tests argument is not recognized in the second case. Please report this to JMP Technical Support (support@jmp.com) so it can be entered as a defect and tracked.

[#3]

This seems to require a bit more work to workaround as "normal" forcing won't work

Names Default To Here(1);

dt = Open("$sample_data/big class.jmp");

x_var = "sex";
y_var = "weight";

myexpr = EvalExpr(dt << Fit Life by X(
	Distribution( Weibull ),
	Nested Model Tests( Location and Scale ),
	Y( :weight ),
	X(Expr(NameExpr(AsColumn(dt, x_var)))),
));
show(myexpr);
obj = Eval(myexpr);

One of the extremely rare cases where I would say to use Eval(Parse()) (evil parse) but this seems to work

Names Default To Here(1);

dt = Open("$sample_data/big class.jmp");

x_var = "sex";
y_var = "weight";

obj2 = Eval(Parse(Eval Insert("\[dt << Fit Life by X(
	Distribution(Weibull),
	Nested Model Tests(Location and Scale),
	Y(:weight),
	X(:^x_var^)
)]\")));

[#4]

This is another workaround as well. Thank you!

[#5 · Accepted solution]

Please see the following.

Because "Relationship" is missing, the codes is trying to make the best guess based on X. But at the place where the codes checks on X, X is not a valid column yet, in this case. So, providing complete information, the codes don't need to walk into the trap.

Names Default To Here(1);

dt = Open( "$sample_data/big class.jmp" );

x_var = "sex";
y_var = "weight";

obj1 = dt << Fit Life by X(
	Distribution( Weibull ),
	Relationship( Location and Scale ),
	Nested Model Tests( Location and Scale ),
	Y( Eval(y_var) ),
	X( :sex ),
);

wait(2);

obj2 = dt << Fit Life by X(
	Distribution( Weibull ),
	Relationship( Location and Scale ),
	Nested Model Tests( Location and Scale ),
	Y( :weight ),
	X( Eval(x_var) ),
);

[#6]

Oh, this is exactly the right answer. Thank you very much!