topic Re: Listbox Graphs in Discussions

Listbox Graphs

trevorphysics — Sun, 11 Jun 2023 11:25:54 GMT

I would like to get a listbox plot similar to the script below but with a different data format. My data looks more like the following table where all of the data to plot is in the first two columns. I want the list portion in the list box to be test count, and when I click on a test count, it produces a plot for both forward and reverse. It looks like it may be possible with a bivariate plot using the by input... Bivariate(I1,V1,By Sweep Type, By Test Count).

Is this possible and could I get some guidance on how to get this working. I am a JMP newbie and I'm not entirely sure where to start.

My Data Format

V1	I1	Test Count	Sweep Type
0	0	1	Forward
1	.1	1	Forward
2	.2	1	Forward
0	0	1	Reverse
-1	-.1	1	Reverse
-2	-.2	1	Reverse
0	0	2	Forward
1	.1	2	Forward
2	.2	2	Forward

Listbox Example from sample data

// Listbox_Graphs.jsl
// Display one graph at a time using a listbox to select which one
names default to here(1);

dt    = open("$sample_data\CrimeData.jmp", invisible); 
xcol  = "Year";
ycols = {"Population", "Total Rate", "Violent Rate", "Property Rate", "Murder Rate",
"Rape Rate", "Robbery Rate", "Agg-Aslt Rate", "Burglary Rate", "Larceny Rate",
"MVTheft Rate", "Murder", "Rape", "Robbery", "Agg-Aslt", "Burglary", "Larceny",
"MVTheft", "Total", "Violent", "Property", "Latitude", "Longitude"};

gb_expr = {};
for(i = 1, i <= nitems(ycols), i++,
	ycol = ycols[i];
// Define an array of graph builder expressions, one for each y column
	one_expr = evalinsert(
"\[gb_expr[^i^] = expr(
		dt << Graph Builder(
			Show Control Panel( 0 ),
			Variables( X( :Name( "^xcol^" ) ), Y( :Name( "^ycol^" ) ) ),
			Elements( Points( X, Y, Legend( 5 ) ), Smoother( X, Y, Legend( 6 ) ) )
		),
	);
]\");


	eval(parse(one_expr));
);

// Code to run when a different y variable is selected
y_list_expr = expr(
	y_mtx = y_lb << get selected indices;
	y_i   = y_mtx[1];
	hlb << delete;
	vlb << append(
		hlb = hlistbox(
			eval(gb_expr[y_i])
		)
	);
);

gw = new window("Graphical Output",
	hlistbox(
		panel box(
			xcol || " vs:",
			y_lb = list box(ycols,
				width(250),
				nlines(20),
				max selected(1),
				y_list_expr,
			)
		),
		panel box("Graph",
			vlb = vlistbox(
				hlb = hlistbox(
					eval(gb_expr[1]),
				)
			)
		)
	)
);

y_lb << set selected(1);

Re: Listbox Graphs

jthi — Mon, 18 Jul 2022 17:06:10 GMT

Maybe using Local Data Filter would be enough? Change Test Count modeling type to Ordinal or Nominal and then you can get something like this:

Script created by JMP:

Graph Builder(
	Size(536, 491),
	Show Control Panel(0),
	Variables(X(:V1), Y(:I1), Group X(:Test Count), Group X(:Sweep Type)),
	Elements(Points(X, Y, Legend(46)), Smoother(X, Y, Legend(47))),
	Local Data Filter(
		Add Filter(
			columns(:Test Count),
			Where(:Test Count == 1),
			Display(:Test Count, N Items(2), "List Display")
		)
	)
)

Edit:

Script with example data table

Names Default To Here(1);

dt_example = New Table("Untitled",
	Add Rows(9),
	New Column("V1", Numeric, "Continuous", Format("Best", 12), Set Values([0, 1, 2, 0, -1, -2, 0, 1, 2])),
	New Column("I1", Numeric, "Continuous", Format("Best", 12), Set Values([0, 0.1, 0.2, 0, -0.1, -0.2, 0, 0.1, 0.2])),
	New Column("Test Count", Numeric, "Continuous", Format("Best", 12), Set Values([1, 1, 1, 1, 1, 1, 2, 2, 2])),
	New Column("Sweep Type",
		Character(16),
		"Nominal",
		Set Values({"Forward", "Forward", "Forward", "Reverse", "Reverse", "Reverse", "Forward", "Forward", "Forward"})
	)
);

gb = dt_example << Graph Builder(
	Size(536, 491),
	Show Control Panel(0),
	Variables(X(:V1), Y(:I1), Group X(:Test Count), Group X(:Sweep Type)),
	Elements(Points(X, Y, Legend(46)), Smoother(X, Y, Legend(47))),
	Local Data Filter(
		Add Filter(
			columns(:Test Count),
			Modeling Type(:Test Count, Nominal),
			Display(:Test Count, N Items(2), "List Display")
		)
	)
);

Re: Listbox Graphs

trevorphysics — Mon, 18 Jul 2022 17:22:47 GMT

This may work. Is there a way to split up the axis? The real data has a much larger range of voltages and currents

Functionally this is a great start. Thanks for your help!

Re: Listbox Graphs

trevorphysics — Mon, 18 Jul 2022 17:31:00 GMT

Edit: Posted the comment twice. Apologies.

Re: Listbox Graphs

jthi — Mon, 18 Jul 2022 17:46:58 GMT

I'm not sure what you mean by splitting the axis. You could possibly create new column with additional grouping and use that as other local data filter or for example as Page

Re: Listbox Graphs

txnelson — Mon, 18 Jul 2022 17:49:06 GMT

You may also want to look into using the Response Screening Platform. It generates a summary data table that has the analysis results in significance order. Selection of the 1 or more of the parameters, and then running one of the provided Table Scripts produces the scatterplots using Fit Y by X. Therefore, all of the analytical tests available in Fit Y by X are available to the selected parameters.

Names Default To Here( 1 );
dt = Current Data Table();

Response Screening( 
	Y( :V1, :I1 ), 
	X( :Test Count, :Sweep Type ), 
	PValues Table on Launch( 1 ) 
	);

Re: Listbox Graphs

trevorphysics — Mon, 18 Jul 2022 19:34:50 GMT

Setting it as a different page worked perfectly. Thank you!