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; } } }

Calculating and Plotting Percent of Increase

rmiyasato

Hi. New to JMP.

I am plotting resistance values over time from a csv dump. How do I calculate and plot the percent of increase or delta? Tab or script.

I am collecting data from 99 boards and reading 5 channels per board over time (10x).

I am able to plot the resistance data over time.

Table looks like this:

Graph: Shows resistance values on the left. I want to plot the percent of change increase or decrease.

Is it possible to do this?

Thanks!

Accepted answers

cwillden

Is it possible to do this?

@rmiyasato, Welcome to the community! And yes, anything is possible if you believe in yourself and work hard to achieve your dreams.

What isn't clear is if you want to compute the percentage difference from T00 for all subsequent time points, or the percentage difference from the previous time point. I'll describe both.

Start by sorting your data by BOARD, CHANNEL, and T(x) in that order using Tables > Sort.

% Difference from Previous Time Point

Add a new column with this formula: (you can copy and paste right into the table)

If( :Name( "T(X)" ) != "T00",
	(:Data[Row()] - :Data[Row() - 1]) / :Data[Row() - 1],
	0
)

This uses the row() function to compute the difference of Data in the current row from the previous row, and divide the difference by the value of the previous row.

Format the column to have a Percent format via Column Info.

% Difference from T00

For this, we'll create 2 new columns. The first is purely auxiliary and will contain the T00 value for the current Board/Channel combination. I called the column "T00 Value" and used this column formula:

If( :Name( "T(X)" ) == "T00",
	:Data,
	:T00 Value[Row() - 1]
);

It has a recursive reference to itself, which is normally bad practice.

Next, we'll compute the %Difference from T00 in another new column using this column formula:

(:Data - :T00 Value) / :T00 Value

Change the format to a Percent format.

Here's what the result looks like with my made up table:

I'm attaching this table to this post.

**Note: I should add that if you resort your table again, all the formulas will re-evaluate and screw everything up. You can just remove the formulas so the values are static.

PercentDiffTable.jmp

All comments

Kevin_Anderson

Hi, rmiyasato!

Welcome. Of course, this is possible.

% change is, of course, dependent upon the specification of values in the numerator and denominator of the equation.

Do you wish % change from starting values? From each subsequent value?

rmiyasato

Awesome!!!

From the starting value.

rmiyasato

Here's my current script.

PROBEDATA = Open(
Pick File(),
columns(
Column( "SCANID", Character, "Nominal" ),
Column( "IM#", Character, "Nominal" ),
Column( "LREQ", Numeric, "Continuous", Format( "Best", 10 ) ),
Column(
"DATE/TIME",
Numeric,
"Continuous",
Format( "m/d/y h:m:s", 23, 0 ),
Input Format( "m/d/y h:m:s", 0 )
),
Column( "T(x)", Character, "Nominal" ),
Column( "BOARD", Numeric, "Continuous", Format( "Best", 10 ) ),
Column( "CHANNEL", Numeric, "Continuous", Format( "Best", 10 ) ),
Column( "DATA", Numeric, "Continuous", Format( "Best", 10 ) )
),
Import Settings(
End Of Line( CRLF, CR, LF ),
End Of Field( Comma, CSV( 0 ) ),
Strip Quotes( 1 ),
Use Apostrophe as Quotation Mark( 0 ),
Scan Whole File( 1 ),
Treat empty columns as numeric( 0 ),
CompressNumericColumns( 0 ),
CompressCharacterColumns( 0 ),
CompressAllowListCheck( 0 ),
Labels( 1 ),
Column Names Start( 1 ),
Data Starts( 2 ),
Lines To Read( "All" ),
Year Rule( "20xx" )
)
);

PROBEDATA << Graph Builder(
Size( 1750, 1245 ),
Automatic Recalc( 0 ),
Variables(
X( :Name( "T(x)" ) ),
Y( :DATA ),
Group X(
:CHANNEL,
Order By( :BOARD, Ascending, Order Statistic( "Mean" ) )
),
Group Y( :BOARD, Levels( 99 ), N View Levels( 10 ) )
),
Elements(
Points( X, Y, Legend( 53 ) ),
Caption Box(
X,
Y,
Legend( 54 ),
Summary Statistic( "Range" ),
X Position( "Left" )
)
)
)

cwillden

Is it possible to do this?

@rmiyasato, Welcome to the community! And yes, anything is possible if you believe in yourself and work hard to achieve your dreams.

What isn't clear is if you want to compute the percentage difference from T00 for all subsequent time points, or the percentage difference from the previous time point. I'll describe both.

Start by sorting your data by BOARD, CHANNEL, and T(x) in that order using Tables > Sort.

% Difference from Previous Time Point

Add a new column with this formula: (you can copy and paste right into the table)

If( :Name( "T(X)" ) != "T00",
	(:Data[Row()] - :Data[Row() - 1]) / :Data[Row() - 1],
	0
)

This uses the row() function to compute the difference of Data in the current row from the previous row, and divide the difference by the value of the previous row.

Format the column to have a Percent format via Column Info.

% Difference from T00

If( :Name( "T(X)" ) == "T00",
	:Data,
	:T00 Value[Row() - 1]
);

It has a recursive reference to itself, which is normally bad practice.

Next, we'll compute the %Difference from T00 in another new column using this column formula:

(:Data - :T00 Value) / :T00 Value

Change the format to a Percent format.

Here's what the result looks like with my made up table:

I'm attaching this table to this post.

**Note: I should add that if you resort your table again, all the formulas will re-evaluate and screw everything up. You can just remove the formulas so the values are static.

PercentDiffTable.jmp

rmiyasato

Thanks! First time posting a question. This is awesome! Thanks again for the help! Worked great!