Transform
Calculator
A transform function
is a mathematical or logical operation that may be applied to a
variable or group of variables in the dataset. The result of the
function can replace the values of an existing variable or can be stored
as a new variable. You can apply more than one transform sequentially
to the same variable.
Transforms are recorded in the form:
Variable = Expression
where:
- Expression is a formula built from the transform functions, dataset variables (the inputs), and numeric, date/time, or string constants.
- Variable (the output) is the name of one or more variables into which the results of the calculation will be stored. If A is an input to B, then B is said to depend on A. All transforms require exactly one output variable, unless specifically noted otherwise in“Transform Reference”.
If the output variable already exists in the
dataset, the transform expression must use that same variable as
an input. Replace an existing variable's values with something unrelated
is not supported. If any input to a transform is a computed variable,
that transform's output cannot be a raw variable. Some transform functions
require that the output be the same variable as one of the inputs.
Variables with different date/time references
cannot be combined in a single transform (but if a date/time reference
is none, it is exempt from this rule). You can change the date/time
reference of a raw or independent variable (in the Column Configuration
tab of the dataset manifest) but not of a computed variable, which
inherits its date/time from its input variable.
In general, you can combine as many functions
as you wish into a single, complex transform expression. However,
some transform functions cannot be combined with certain others
in a single expression and will result in an error message like
Invalid mix of vector functions with scalar
function
. If this message appears, apply the functions sequentially
instead of combined in a single transform.Provide Feedback