I have two xts-matrices (A and B) were both contain headers in the form of "value1.value2". This label combination is saved for both matrices as characters to avoid R to attach the prefix "X" to the numbers.
A and B are already labeled with their symbol combination.
Now I want to attach matrix B on the right hand-side of A (I used C <- cbind(A,B) or C <- merge.xts(A,B)).
Merging or binding them causes R to attach the X in front of the headers.
It can't be undone by
colnames(C) <- as.character(colnames(C))
How can I get rid of it, if I want to use numbers as headers (please note: it's not a file that I am loading, I am trying to bring to outputs together)?
As next step I want to order the columns in an increasing manner and tried:
C <- C[,sort(colnames(C), decreasing = F)]
But R completely messes up the order and seems to randomly arrange the columns.
I also tried as.data.frame(cbind(A,B)) and sort the dataframe, but it doesn't improve anything.
I am happy to receive valuable suggestions.
Unfortunately, I can't provide a minimum working example.
Thank you very much!
You can rename the colnames based on the colnames from the individual xts objectsor use a substring like substring(colnames(C), 2).
create data sample:
library(xts)
A <- matrix(1:20, ncol = 2)
colnames(A) <- c("10.20", "20.01")
B <- matrix(20:1, ncol = 2)
colnames(B) <- c("10.10", "10.01")
# as xts objects
A <- as.xts(A, order.by = seq.Date(Sys.Date() - 9, Sys.Date(), by = "day"))
B <- as.xts(B, order.by = seq.Date(Sys.Date() - 9, Sys.Date(), by = "day"))
C <- merge(A, B)
# rename colnames to names before merge to get rid of X in front of name
colnames(C) <- c(colnames(A), colnames(B))
# alternative use substring:
# colnames(C) <- substring(colnames(C), 2)
C[, sort(colnames(C), decreasing = F)]
10.01 10.10 10.20 20.01
2020-03-08 10 20 1 11
2020-03-09 9 19 2 12
2020-03-10 8 18 3 13
2020-03-11 7 17 4 14
2020-03-12 6 16 5 15
2020-03-13 5 15 6 16
2020-03-14 4 14 7 17
2020-03-15 3 13 8 18
2020-03-16 2 12 9 19
2020-03-17 1 11 10 20