Robust Scaling With Median and Mad

RobScale is a wrapper function for robust standardization, using median and mad instead of mean and sd.

RobScale(x, center = TRUE, scale = TRUE)

Arguments

x

a numeric matrix(like object).

center

a logical value defining whether x should be centered by the median. Centering is done by subtracting the column medians (omitting NAs) of x from their corresponding columns. If center is FALSE, no centering is done.

scale

a logical value defining whether x should be scaled by the mad. Scaling is done by dividing the (centered) columns of x by their mad. If scale is FALSE, no scaling is done.

Value

the centered, scaled matrix. The numeric centering and scalings used (if any) are returned as attributes "scaled:center" and "scaled:scale"

Author

Andri Signorell <andri@signorell.net>

See also

scale, sweep

Examples

x <- d.pizza$temperature
plot(x=seq_along(x), y=RobScale(x), xlim=c(0,100))
points(x=seq_along(x), y=scale(x), col="red" )