Transform Columns

Each launch window in JMP enables you to create one or more temporary transform columns for use in performing analyses. These transform columns are not part of the source data table and only can be used within the context of the current launch window. Transform columns use formulas or calculations to define the column values. Closing the launch window or the generated report deletes any transform columns.

Each column listed in the Select Columns pane of the launch window includes an icon representing the column’s modeling type (continuous, ordinal, or nominal) and the column name. Right-click on a column name to create a transform column using Transform, Character, Combine, Aggregate, Distributional, Date Time, Row, or Formula to calculate the column’s values.

Right-click options depend on the selected column’s data type and number of columns selected.

Example of Transform Column Menu

Group By

For ordinal and nominal data, specifies the column to use for grouping data. A separate analysis is computed for each level of the specified column.

Notes:

•

The transform column is available only in the current launch window. To make the transform column available outside of the current launch window, right-click the transform column and select Add to Data Table. The transform column is added to the source data table.

•

You can paste a transform column into a Roles box on the launch window. For example, you might copy a transform column from a script. Right-click in the appropriate launch window Cast Selected Columns into Roles box and select Paste. This is an alternative to right-clicking the column in the Select Columns list, selecting the transform, and adding the transform column to a role.

Transform Menu

Select a function from the Transform menu to create a transform column containing the calculations based on the selected function. For details, see the Scripting Index in the Help menu or Transform in the Fitting Linear Models.

Note: You can apply unary functions to multiple columns resulting in multiple transform columns.

Descriptions of the Transform Menu Options
Square Root	Takes the square root of the values of the selected column.
Square	Calculates the square for the selected column values.
Log	Applies the natural logarithm transformation to the selected column.
Exp	Applies the exponential transformation to the selected column.
Log10	Applies the base-10 logarithm transformation to the selected column.
Pow10	Calculates 10 raised to the power of the selected column values.
Cube Root	Calculates the cube root for the selected column values.
Cube	Calculates the cube for the selected column values.
Reciprocal	Calculates the reciprocal (1/column) for the selected column values.
Absolute Value	Calculates the absolute value for the selected column values.
Negation	Calculates the negative for the selected column values.
Arrhenius	Applies the Arrhenius transformation to the variable T (temperature in degrees Centigrade): This is the component of the Arrhenius relationship that is multiplied by the activation energy.
Arrhenius Inverse	Applies the inverse of the Arrhenius transformation to the variable X:
Logit	Calculates the inverse of the logistic function for the selected column (where p is in the range of 0 to 1):
Logistic	Calculates the logistic (also known as Squish and Logist) function for the selected column (where the result is in the range of 0 to 1):
Logit Percent	Calculates the logit as a percent for the selected column (where pct is a percent in the range of 0 to 100):
Logistic Percent	Calculates the logistic (or logist) as a percent for the selected column (where the result is in the range of 0 to 100):

Combine Menu

Select multiple columns to access the Combine menu. The Combine menu creates a transform column containing the calculations based on the selected function. For details, see the Scripting Index in the Help menu.

The following functions are included in the menu:

Sum

Calculates the sum of the first and second columns (A + B).

Difference

Calculates the difference between the first and second columns (A - B).

Difference (reverse order)

Calculates the difference between the second and first columns (B - A).

Product

Calculates the product of the first and second columns (A X B).

Ratio

Calculates the ratio of the first column to the second column (A / B).

Ratio (reverse order)

Calculates the ratio of the second column to the first column (B / A).

Minimum

Returns the minimum value of the selected columns.

Maximum

Returns the maximum value of the selected columns.

Average

Returns the average value of the selected columns.

Aggregate Menu

Select a function from the Aggregate menu to create a transform column containing the statistics calculated from the selected column (or part of a column if you specified a Group By column). For details, see the Scripting Index in the Help menu.

Note: The Group By option is useful for these functions.

The following functions are included in the menu:

Mean

Returns the average value of the selected column.

Sum

Calculates the sum of the values in the selected column.

Count

Calculates the number of values in the selected column.

Median

Calculates the median value for the selected column.

Quantile

Calculates the quantile of the specified percentage for the selected column.

Minimum

Returns the minimum value of the selected column.

Maximum

Returns the maximum value of the selected column.

Standard Deviation

Calculates the standard deviation of the values in the selected column.

Distributional Menu

Select a function from the Distributional menu to create a transform column containing the statistics calculated from the selected column. For details, see the Scripting Index in the Help menu.

The following functions are included in the menu:

Center

Subtracts the column mean from each value across all rows of the selected column.

Standardize

Calculates the column mean divided by the standard deviation across all rows of the selected column.

Range 0 to 1

Scales the data up or down so that the minimum value is greater or equal to 0, and the maximum value is less than or equal to 1.

Box Cox

Transforms the data using the Box-Cox equation. For details, see Box Cox Y Transformation in the Fitting Linear Models.

Johnson Normalizing

Transforms the data using one of the Johnson equations. The new column name indicates either Johnson Su, Johnson Sb, or None, depending on which equation was used to calculate the new data.

Informative Missing

Creates two columns. The Informative column replaces missing values with the column mean. The Is Missing column indicates 1 for missing values, and 0 otherwise.

Date Time Menu

For column values containing date or time values, select a function from the Date Time menu to create a transform column containing values calculated from the selected column. For details, see the Scripting Index in the Help menu.

The following functions are included in the menu:

Day

Returns the day of the month for the date in the selected column.

Month

Returns the month number for the date in the selected column.

Month Abbr.

Returns the abbreviated month for the date in the selected column.

Year

Returns the year for the date in the selected column.

Month Year

Returns the month number and year for the date in the selected column.

Quarter