Dear Rui,
Many thanks for your response.
Best,
SV
Le lundi 3 août 2020 à 16:54:35 UTC+2, Rui Barradas <ruipbarra...@sapo.pt> a
écrit :
Hello,
No, there isn't a built-in that I know of.
Here is one:
double.mad <- function(x, include.right = FALSE, na.rm = FALSE){
if(na.rm) x <- x[!is.na(x)]
m <- median(x)
odd <- (length(x) %% 2L) == 1L
out <- if(odd){
if(include.right) {
c(lo = mad(x[x < m]), hi = mad(x[x >= m]))
} else {
c(lo = mad(x[x <= m]), hi = mad(x[x > m]))
}
} else {
c(lo = mad(x[x < m]), hi = mad(x[x > m]))
}
out
}
double.mad(x)
# lo hi
#0.81543 0.44478
double.mad(c(x, 1))
# lo hi
#2.29803 0.44478
double.mad(c(x, 1), include.right = TRUE)
# lo hi
#1.03782 1.63086
Hope this helps,
Rui Barradas
Às 15:22 de 03/08/2020, varin sacha via R-help escreveu:
> Dear R-Experts,
>
> Is there an all-ready function to calculate the Double MAD (Median absolute
> deviation) as there is an easy function to calculate the MAD "mad function".
> Or I have to write my own function for Double MAD ?
>
> To calculate the double MAD, the idea is the following : for the obtained
> median value, we should calculate two median absolution deviations. One
> deviation should be calculated for the numbers below the median and one for
> the numbers above the median:
>
> Here is the very easy reproducible example :
>
> x<-c(2.5,4.4,3.2,2.1,1.3,2.6,5,6.6,5,5,6.1,7.2,9.4,6.9)
> mad(x)
>
> ______________________________________________
> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
--
Este e-mail foi verificado em termos de vírus pelo software antivírus Avast.
https://www.avast.com/antivirus
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
