Question

Here is data:

myd <- data.frame (var1.1 = rnorm (10, 5,2), var1.2 = rnorm (10, 3,1),
                   varA.1 = rnorm (10, 5,2), varA.2 = rnorm (10,1,2), 
                    C.1 = rnorm (10, 3,1), C.2 = rnorm (10, 4,1))

I want create a list with of size 3 (n/2, if n is number of variable in dataframe) (with two columns each)

mylist <- as.list (myd)

creats a list with each variable as component. But I want create list of pair of variable in one unit – var1.1 and var1.2, similarly varA.1 and varA.2 or C.1 and C.2. Thus new list unit has 2 columns each. Any solution ?

Answer 1

I think this will give you what you want:

lapply(seq(2, ncol(myd), by = 2), function(i) myd[, (i-1):i])

Now lets say you wanted to name each element in the list by the shared variable names of the data frame:

X <- lapply(seq(2, ncol(myd), by = 2), function(i) myd[, (i-1):i])
names(X) <- sapply(X, function(x){
    strsplit(names(x)[1], '\\.')[[c(1, 1)]]
})
X