topic Re: How to add Kurtosis and Skewness to Tabulate in Discussions

How to add Kurtosis and Skewness to Tabulate

JMPPRO — Sat, 10 Jun 2023 23:26:41 GMT

Hello,

I would like to add "Kurtosis" and "Skewness" to my row table. How do I do that?

dt = Current Data Table();
tab = dt << Tabulate(
    Show Control Panel( 0 ),
    Add Table(
        Column Table( Analysis Columns( :VAR1, :VAR2, :VAR3, ) ),
        Row Table( Statistics( Mean, Std Dev, Min, Max, Range, CV, Median ) )
    )
);
Wait( 0.1 );
dtTab = tab << makeIntoDataTable;
Write( "Make a new data table from the Tabulate Platform", "\!N" );

Re: How to add Kurtosis and Skewness to Tabulate

jthi — Mon, 01 Mar 2021 19:53:56 GMT

I think you cannot get kurtosis and skewness through tabulate but they can be found from Distribution platform.

Create Distribution with only distributions with the variables you are interested in
Hold CTRL (this will make the options affect all of the distributions) and click red triangle in one of the distributions of the variables
Go to Display Options / Customize Summary Statistics -> select statistics you are interested in
Right click on one of the summary statistics tables in Distribution platform and choose Make Combined Data Table
Split result data table if needed

Example script with Semiconductor Capability sample data (scripts copy pasted almost directly from JMP):

Names Default To Here(1);
dt = Open("$SAMPLE_DATA/Semiconductor Capability.jmp");

platform = dt <<
Distribution(
	Continuous Distribution(
		Column(:NPN1),
		Summary Statistics(1),
		Customize Summary Statistics(
			Std Err Mean(0),
			Upper Mean Confidence Interval(0),
			Lower Mean Confidence Interval(0),
			N(0),
			Skewness(1),
			Kurtosis(1),
			CV(1),
			Minimum(1),
			Maximum(1),
			Median(1)
		),
		Process Capability(0)
	),
	Continuous Distribution(
		Column(:PNP1),
		Summary Statistics(1),
		Customize Summary Statistics(
			Std Err Mean(0),
			Upper Mean Confidence Interval(0),
			Lower Mean Confidence Interval(0),
			N(0),
			Skewness(1),
			Kurtosis(1),
			CV(1),
			Minimum(1),
			Maximum(1),
			Median(1)
		),
		Process Capability(0)
	),
	Continuous Distribution(
		Column(:PNP2),
		Summary Statistics(1),
		Customize Summary Statistics(
			Std Err Mean(0),
			Upper Mean Confidence Interval(0),
			Lower Mean Confidence Interval(0),
			N(0),
			Skewness(1),
			Kurtosis(1),
			CV(1),
			Minimum(1),
			Maximum(1),
			Median(1)
		),
		Process Capability(0)
	),
	Histograms Only
);
Wait(0);
dt_statistics = Report(platform)[Outline Box("Distributions")][Outline Box("NPN1")][
Outline Box("Summary Statistics")][Table Box(1)] << Make Combined Data Table;
Report(platform) << Close Window;
Close(dt, no save);
dt_split = dt_statistics << Split(
	Split By(:Column 1),
	Split(:Column 2),
	Group(:Y),
	Sort by Column Property
);
Close(dt_statistics, no save);

Re: How to add Kurtosis and Skewness to Tabulate

Byron_JMP — Tue, 02 Mar 2021 13:07:11 GMT

That's really a pretty good question, maybe reframing it another way..

Rather than combining statistics, using complex scripting, wouldn't it be easier if Tabulate included more, or maybe all the univariate statistics included in the distribution platform?

..Yes, yes it would.

I'm staff, so I can't do this, but you could add this to the Wish List (its in blue menu bar at the top of the window.)

Re: How to add Kurtosis and Skewness to Tabulate

hogi — Sun, 06 Jul 2025 15:54:47 GMT

@Byron_JMP wrote:

wouldn't it be easier if Tabulate included more, or maybe all the univariate statistics included in the distribution platform?

besides Tabulate, the same holds for:

- Tables/Summary
- and the Summary Statistics in Graph Builder!

I submitted the idea to synchronize the statistics options of the different platforms Together with

The original title was something like Col N Categories - and all the others ...

Col N Categories will come with JMP 19 (with the name Col N Unique()) - so, the wish got marked as "delivered".

"for all the others" - here is a new wish:

unfortunately, it starts again with 0 Kudos : (

Still a long time till it gets the chance to get into a future version of JMP - but maybe somebody sees this post and says:

I'm staff, [and] I can[] do this.