3

There are multiple XTS objects. Each one contains one column only (of prices). The objects may be of different lengths. I want to merge them by index. I can merge two XTS objects with merge.xts. But I want to merge multiple (with one function, without repeating merging process multiple times).

So I applied the approach recommended here. It works. Just one issue: it merges all data into one column, rather than placing each XTS object to be merged into its own column. e.g. In the reproducible example, the desired end result is for XTSM to have 3 columns.

How can that be achieved? Thank you.

library('xts')

#Data
XTS1 <- structure(c(125.26, 125.14, 125.21, 125.26, 125.21, 125.26, 125.24, 125.22, 125.3, 125.32, 125.38, 125.38, 125.29, 125.26, 125.3, 125.25, 125.29, 125.37, 125.21, 125.3, 125.25, 125.24, 125.21, 125.24, 125.24, 125.26, 125.23, 125.23), class = c("xts", "zoo"), .indexCLASS = c("POSIXct", "POSIXt"), tclass = c("POSIXct", "POSIXt"), .indexTZ = "", tzone = "", .CLASS = "double", index = structure(c(1403537400, 1403538300, 1403539200, 1403540100, 1403541000, 1403541900, 1403542800, 1403543700, 1403544600, 1403545500, 1403546400, 1403547300, 1403548200, 1403549100, 1403550000, 1403550900, 1403551800, 1403552700, 1403553600, 1403554500, 1403555400, 1403556300, 1403557200, 1403560800, 1403561700, 1403562600, 1403563500, 1403564400), tzone = "", tclass = c("POSIXct", "POSIXt")), .Dim = c(28L, 1L))
XTS2 <- structure(c(1867, 1867, 1868.5, 1868.75, 1868.75, 1868.75, 1868.25, 1867.25, 1867.5, 1867.75, 1868, 1868.25, 1868.25, 1868.75, 1868, 1868.5, 1869, 1869.75, 1868.75, 1868.5, 1868.25, 1867.5, 1867, 1866.75, 1866.75, 1867, 1867.5), class = c("xts", "zoo"), .indexCLASS = c("POSIXct", "POSIXt"), tclass = c("POSIXct", "POSIXt"), .indexTZ = "", tzone = "", .CLASS = "double", index = structure(c(1403537400, 1403538300, 1403539200, 1403540100, 1403541000, 1403541900, 1403542800, 1403543700, 1403544600, 1403545500, 1403546400, 1403547300, 1403548200, 1403549100, 1403550000, 1403550900, 1403551800, 1403552700, 1403553600, 1403555400, 1403556300, 1403557200, 1403560800, 1403561700, 1403562600, 1403563500, 1403564400), tzone = "", tclass = c("POSIXct", "POSIXt")), .Dim = c(27L, 1L))
XTS3 <- structure(c(1867, 1867, 1868.5, 1868.75, 1868.75, 1868.75, 1868.25, 1867.25, 1867.5, 1867.75, 1868, 1868.25, 1868.25, 1868.75, 1868, 1868.5, 1869, 1869.75, 1868.75, 1868.5, 1868.25, 1867.5, 1867, 1866.75, 1866.75, 1867, 1867.5), class = c("xts", "zoo"), .indexCLASS = c("POSIXct", "POSIXt"), tclass = c("POSIXct", "POSIXt"), .indexTZ = "", tzone = "", .CLASS = "double", index = structure(c(1403537400, 1403538300, 1403539200, 1403540100, 1403541000, 1403541900, 1403542800, 1403543700, 1403544600, 1403545500, 1403546400, 1403547300, 1403548200, 1403549100, 1403550000, 1403550900, 1403551800, 1403552700, 1403553600, 1403555400, 1403556300, 1403557200, 1403560800, 1403561700, 1403562600, 1403563500, 1403564400), tzone = "", tclass = c("POSIXct", "POSIXt")), .Dim = c(27L, 1L))

#Following code merges 2 xts objects into a 2 column xts
XTSM <- merge(XTS1,XTS2, all=TRUE)

#Following code merges multiple xts objects into a 1 column xts
XTSlist <- list(XTS1, XTS2, XTS3)

do.call.rbind <- function(lst) {
  while(length(lst) > 1) {
    idxlst <- seq(from=1, to=length(lst), by=2)
    lst <- lapply(idxlst, function(i) {
      if(i==length(lst)) { return(lst[[i]]) }
      return(rbind(lst[[i]], lst[[i+1]]))
    })
  }
  lst[[1]]
}

XTSM <- do.call.rbind (XTSlist)
Joshua Ulrich
  • 173,410
  • 32
  • 338
  • 418
Krug
  • 1,003
  • 13
  • 33
  • 3
    `merge.xts` accepts more than 2 xts objects. See `?merge.xts` – G. Grothendieck Jul 04 '17 at 17:27
  • Indeed it does. That solved the issue. For some odd reason when I tried multiple objects with `merge.xts` before I was getting an error message. Thanks. – Krug Jul 04 '17 at 17:33
  • 2
    Note that `merge.xts` only accepts two arguments if `join` is not `"all"`. Also note that you should just call `merge()`. It is not good practice to call methods directly. There's nothing stopping you from calling, for example `merge.xts(data.frame(...))`, which could have unexpected results. – Joshua Ulrich Jul 04 '17 at 17:42
  • 1
    `mege.zoo` accepts multiple arguments in all cases. – G. Grothendieck Jul 04 '17 at 22:34

1 Answers1

2

Joshua answer the question in the comments. Posting solution for the sake of having the Q answered. It is as simple as this:

XTSM.T <- merge(XTS1,XTS2,XTS3, all=TRUE)

or

XTSM.F <- merge(XTS1,XTS2,XTS3, all=FALSE)
Krug
  • 1,003
  • 13
  • 33