aad: Average Absolute Deviation
In abnormally-distributed/cvreg: Cross Validation and Robust Estimation Utilities

The average absolute deviation (AAD) from the median is an alternative to the median absolute deviation (MAD). The AAD is in fact the scale parameter for a double exponential (Laplacian) distribution in just the same way that the standard deviation is the scale parameter for the Gaussian distribution. An advantage that the AAD does not break down when 50% of the values are the same like the MAD. Like the MAD, the AAD must be scaled by a constant to serve as an estimator of the standard deviation. Here, the product of the raw AAD and the the reciprocal of sqrt(2/pi), or 1.253314, yields a consistent estimator of the standard deviation.

1	aad(x, constant = 1.253314, center = hdmedian)

`x`	a numeric vector
`constant`	the constant scaling factor. defaults to 1.253314.
`center`	a function to use for estimating the center, from which the average absolute deviations are measured. a number may also be provided. the default is to use the Harrell-Davis estimate of the median.