I am trying to obtain the column MAD my nested list and return my nested list with the same structure. My example of list as below from this page
k = 5
library(mvtnorm)
set.seed(123)
listofdfs <- list()
a=0
variance=1
g1=5
g2=5
mn=0
covar=matrix(c(variance,a,a,a,variance,a,a,a,variance),ncol=3)
for (i in 1:k){
x1 = rmvnorm(n=g1,mean=c(mn,mn,0),sigma=covar)
x2 = rmvnorm(n=g2,mean=c(mn,mn,0),sigma=covar)
data = list(x1=x1,x2=x2) # to make each sub element to x1 and x2
df = data
listofdfs[[i]]=df
}
listofdfs
#rename the datasets
listofdfs <- setNames(listofdfs, paste0('df', 1:5))
head(listofdfs)
$df1
$df1$x1
[,1] [,2] [,3]
[1,] -0.56047565 -0.2301775 1.5587083
[2,] 0.07050839 0.1292877 1.7150650
[3,] 0.46091621 -1.2650612 -0.6868529
[4,] -0.44566197 1.2240818 0.3598138
[5,] 0.40077145 0.1106827 -0.5558411
[6,] 1.78691314 0.4978505 -1.9666172
[7,] 0.70135590 -0.4727914 -1.0678237
[8,] -0.21797491 -1.0260044 -0.7288912
[9,] -0.62503927 -1.6866933 0.8377870
[10,] 0.15337312 -1.1381369 1.2538149
$df1$x2
[,1] [,2] [,3]
[1,] 0.42646422 -0.29507148 0.8951257
[2,] 0.87813349 0.82158108 0.6886403
[3,] 0.55391765 -0.06191171 -0.3059627
[4,] -0.38047100 -0.69470698 -0.2079173
[5,] -1.26539635 2.16895597 1.2079620
[6,] -1.12310858 -0.40288484 -0.4666554
[7,] 0.77996512 -0.08336907 0.2533185
[8,] -0.02854676 -0.04287046 1.3686023
[9,] -0.22577099 1.51647060 -1.5487528
[10,] 0.58461375 0.12385424 0.2159416
I want to find the value of MAD for each column in x1 and x2 for k=5. I'm using this function and obtain value MAD for column 1 of x1 from $df1 to $df5:
lapply(listofdfs, function(y) mad(y$x1[,1]))
$df1
[1] 0.6720464
$df2
[1] 0.6111537
$df3
[1] 0.5019855
$df4
[1] 0.7117195
$df5
[1] 0.6085593
I have to use this function to obtain the value of MAD the rest of the column in both x1 and x2. How can I obtain the value of MAD (for k=5 in column =3 for both x1 and x2) in one single function?
It sounds like you're looking for something like this:
library(purrr)
map(listofdfs, # for each element in listofdfs
\(d) map(d, # for each matrix in that element
\(x) map(ncol(x) |> seq_len(), # for each column
\(n) stats::mad(x[,n])))) # calculate mad
Output:
[[1]]
[[1]]$x1
[[1]]$x1[[1]]
[1] 0.5788186
[[1]]$x1[[2]]
[1] 0.5053593
[[1]]$x1[[3]]
[1] 1.551788
[[1]]$x2
[[1]]$x2[[1]]
[1] 0.8124393
[[1]]$x2[[2]]
[1] 0.8201937
[[1]]$x2[[3]]
[1] 1.835052
[[2]]
[[2]]$x1
[[2]]$x1[[1]]
[1] 0.6696449
[[2]]$x1[[2]]
[1] 0.9381823
[[2]]$x1[[3]]
[1] 0.7699464
[[2]]$x2
[[2]]$x2[[1]]
[1] 0.9090718
[[2]]$x2[[2]]
[1] 0.247186
[[2]]$x2[[3]]
[1] 1.012018
[[3]]
[[3]]$x1
[[3]]$x1[[1]]
[1] 0.8265921
[[3]]$x1[[2]]
[1] 0.3067163
[[3]]$x1[[3]]
[1] 0.9440248
[[3]]$x2
[[3]]$x2[[1]]
[1] 0.3024162
[[3]]$x2[[2]]
[1] 0.1273361
[[3]]$x2[[3]]
[1] 0.7478695
[[4]]
[[4]]$x1
[[4]]$x1[[1]]
[1] 1.156026
[[4]]$x1[[2]]
[1] 1.32832
[[4]]$x1[[3]]
[1] 0.703393
[[4]]$x2
[[4]]$x2[[1]]
[1] 0.8368312
[[4]]$x2[[2]]
[1] 1.234974
[[4]]$x2[[3]]
[1] 0.7706332
[[5]]
[[5]]$x1
[[5]]$x1[[1]]
[1] 0.1668533
[[5]]$x1[[2]]
[1] 1.520311
[[5]]$x1[[3]]
[1] 0.4594677
[[5]]$x2
[[5]]$x2[[1]]
[1] 1.408639
[[5]]$x2[[2]]
[1] 0.1901449
[[5]]$x2[[3]]
[1] 0.3374949