Weighted Median Absolute Deviation (MAD)
Computes a weighted MAD of a numeric vector.
weightedMad(x, w = NULL, idxs = NULL, na.rm = FALSE, constant = 1.4826, center = NULL, ...) rowWeightedMads(x, w = NULL, rows = NULL, cols = NULL, na.rm = FALSE, constant = 1.4826, center = NULL, ...) colWeightedMads(x, w = NULL, rows = NULL, cols = NULL, na.rm = FALSE, constant = 1.4826, center = NULL, ...)
x |
a |
w |
a vector of weights the same length as |
idxs, rows, cols |
A |
na.rm |
a logical value indicating whether |
constant |
|
center |
Optional |
... |
Not used. |
Returns a numeric scalar.
Missing values are dropped at the very beginning,
if argument na.rm is TRUE, otherwise not.
Henrik Bengtsson
For the non-weighted MAD, see mad. Internally
weightedMedian() is used to calculate the weighted median.
x <- 1:10 n <- length(x) m1 <- mad(x) m2 <- weightedMad(x) stopifnot(identical(m1, m2)) w <- rep(1, times = n) m1 <- weightedMad(x, w) stopifnot(identical(m1, m2)) # All weight on the first value w[1] <- Inf m <- weightedMad(x, w) stopifnot(m == 0) # All weight on the first two values w[1:2] <- Inf m1 <- mad(x[1:2]) m2 <- weightedMad(x, w) stopifnot(identical(m1, m2)) # All weights set to zero w <- rep(0, times = n) m <- weightedMad(x, w) stopifnot(is.na(m))
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.