nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Standard deviation Bars

Johnanderson

Hello,

I would like to plot Y by X continous variables. X represent unit time., and Y is mean signal at each time point in X (please see below). I would like to add standard deviation bars to each point. How can I do it in JMP pro 14?

Fit Y by X.png

Your help and advices are highly appreciated!

john

Accepted answers

txnelson

Here is the graph

error bars2.PNG

I Stacked the columns

Tables==>Stack

Choosing the 3 Signal columns, Selecting to Keep the Time column, and changing the names of the output columns from Label to Group and Data to Signal

Then using Graph Builder, I dragged Signal to Y, Time to X. I changed the Summary Statistic to Mean and then set the Error Bars to Standard Error.

Next I Right Clicked on the graph and selected Add==>Points

Below is a script that performs these tasks in JSL

Names Default To Here( 1 );
dt = Data Table( "jmp_signals" );
dtStack = dt <<
Stack(
	columns( :Signal 2, :Signal 3, :Signal ),
	Source Label Column( "Group" ),
	Stacked Data Column( "Signal" )
);

dtStack << Graph Builder(
	Variables( X( :time ), Y( :Signal ), Color( :Group ) ),
	Elements(
		Points(
			X,
			Y,
			Legend( 10 ),
			Summary Statistic( "Mean" ),
			Error Bars( "Standard Error" )
		),
		Points( X, Y, Legend( 11 ), Jitter( "None" ) )
	)
);

All comments

txnelson

You can easily handle this using Graph Builder

error bars.PNG

Johnanderson

Dear txnelson – thank you so much for the response! I think you were able to show the standard deviation bars because the modeling type for X (i.e. age) is ordinal not continous. Correct? Is there any differnce?

Kindly, how happened that you have multiple heights for each age group. I expect that your spreadsheet has three columns one for age, the other one for height, and the third one for weight?

Also the title of your figure was "mean (height) & height vs age". That means you ploted two Y variables against one X variable (i.e. age). Correct?

I coud not do it! In my spreadsheet I have only two columns one for X and the other one for Y. both are continous variables. I highly appreciate any guidance. Thanks!

txnelson

Where are you expecting that you are going to get your standard deviation? If you X variable is continuious.....each individual point will be considered independently.

You can easily change the Modeling Type from Continuous to Ordinal, if you have multiple values at the differenx X's

Johnanderson

Dear txnelson – I tried to change X to ordinal. When ploting X vs Y this will output one point in X for each point in Y. In the figure you provided there are multiple points in Y for each X. Kindly, why?