rescale {lessR}

Rescale a Variable

Description

Rescale a variable to either z-scores with a mean of 0 and standard deviation of 1, normalized with a minimum of 0 and a maximum of 1, or to a variable computed like a z-score except use the median in place of the mean and the IQR in place of the standard deviation.

Usage

rescale(x, data=d, kind="z", digits_d=3)

Arguments

x

Variable to rescale.

data

Data frame that contains x.

kind

Type of rescaling.

digits_d

Number of significant digits.

Details

The default rescaling is standardization to z-scores, explicit with kind set to "z", or just centering about the mean with "center". For the min-max normalization to a range from 0 to 1, set kind to "0to1". For the robust equivalent of standardization, set kind to "robust".

If x is a vector in the global environment, then set data to NULL.

Value

The rescaled data.

Author(s)

David W. Gerbing (Portland State University; [e-mail])

See Also

scale.

Examples

# z-score for m01 
d <- Read("Employee")

d[, .("Salary")] 
x <- rescale(Salary)
x