JMP Wish List

francois_berger

What inspired this wish list request?

For DOE analysis (Fit Model) people do not understand why the blue bar is for a pvalue threshold at 0.01 (and not 0.05)

(in DOE, small sample, not a lot of factors)

What is the improvement you would like to see?

Can we set the blue bar with a default pvalue threshold at 0.05?

And can we have an option to change it?

Why is this idea important?

People sometimes remove significant effects (e.g. pvalue=0.03, below the blue bar)

MathStatChem

I assume you are talking about the reference line that used in the plot of LogWorth by Effect in the Effect Summary. It is set at a fixed value of 2, which corresponds to a log worth of -Log10(0.01).

You can change the reference line with JSL, by sending the message << Reference line(value) to that display box (a plot col box).

Example:

names default to here(1);

dt=Open( "$SAMPLE_DATA/Design Experiment/Reactor 20 Custom.jmp" );

fm=dt << Fit Model(
	Y( :Percent Reacted ),
	Effects(
		:Catalyst, :Temperature, :Concentration, :Catalyst * :Temperature,
		:Temperature * :Concentration
	),
	Personality( "Standard Least Squares" ),
	Emphasis( "Effect Screening" ),
	Run(
		:Percent Reacted << {Summary of Fit( 0 ), Analysis of Variance( 0 ),
		Parameter Estimates( 1 ), Effect Details( 0 ), Sorted Estimates( 0 ),
		Plot Actual by Predicted( 1 ), Plot Regression( 0 ),
		Plot Residual by Predicted( 1 ), Plot Studentized Residuals( 1 ),
		Plot Effect Leverage( 0 ), Plot Residual by Normal Quantiles( 0 ),
		Box Cox Y Transformation( 0 ), Profiler(
			1,
			Confidence Intervals( 1 ),
			Desirability Functions( 1 ),
			Term Value(
				:Catalyst( 1.5, Lock( 0 ), Show( 1 ) ),
				:Temperature( 160, Lock( 0 ), Show( 1 ) ),
				:Concentration( 4.5, Lock( 0 ), Show( 1 ) )
			)
		)}
	)
);

pcb=report(fm)["Effect Summary"][Plot Col Box(1)];

// set the reference line to the logWorth equivalent to a p-value of 0.0001

pcb << set reference line(-log10(0.0001));

Valerie_Nedbal

Hello @francois_berger , thank you for your request. We will discuss this wish internally and keep you informed. And thanks @MathStatChem for your workaround.

JMP Wish List

pvalue threshold in Fit Model