I am trying to accomplish the following task to get to matrix d:
d1<-matrix(as.factor(rep(sample(1:10,10,T),5)),ncol=5)
d2<-matrix(as.factor(rep(sample(1:10,10,T),5)),ncol=5)
d3<-matrix(as.factor(rep(sample(1:10,10,T),5)),ncol=5)
d<-cbind(
cbind(d1[,2],d1[,5]),
cbind(d2[,2],d2[,5]),
cbind(d3[,2],d3[,5])
)
But for many matrices d1...dn, say.
More generally I would like to select the same column numbers from a series of matrices and append into a single matrix. The focus of this task is on combining, not creating the matrices. The factor-type column vectors should be preserved.
I thought about something along the lines of
d<-matrix(nrow=10)
dl<-list(d1,d2,d3)
for (i in 1:3){
d<-cbind(d,dl[[i]][,2],dl[[i]][,5])
}
But maybe there is a better way.
You can create a list of your matrices and use do.call and lapply to get what you want:
matList <- list(d1, d2, d3)
do.call(cbind, lapply(matList, function(x) x[, c(2, 5)]))
# [,1] [,2] [,3] [,4] [,5] [,6]
# [1,] "3" "3" "3" "3" "10" "10"
# [2,] "4" "4" "2" "2" "3" "3"
# [3,] "6" "6" "7" "7" "7" "7"
# [4,] "10" "10" "4" "4" "2" "2"
# [5,] "3" "3" "8" "8" "3" "3"
# [6,] "9" "9" "5" "5" "4" "4"
# [7,] "10" "10" "8" "8" "1" "1"
# [8,] "7" "7" "10" "10" "4" "4"
# [9,] "7" "7" "4" "4" "9" "9"
# [10,] "1" "1" "8" "8" "4" "4"
By the way, the data type in your matrix is character, not factor. See the help page at ?matrix where you will find the following:
The method for data frames will return a character matrix if there is only atomic columns and any non-(numeric/logical/complex) column, applying as.vector to factors and format to other non-character columns.