Union of Time Series
DESCRIPTION:
Binds several time series together into a single multivariate time series whose time domain is the union of the time domains of the component series.
USAGE: 
ts.union(..., dframe = F, eps = .Options$ts.eps)
REQUIRED ARGUMENTS:
...:
any number of arguments, each of which must be a time series of class "rts", "cts", or "its".
OPTIONAL ARGUMENTS:
dframe:
logical flag: if TRUE the resulting time series is a data frame. The default, FALSE, returns a matrix.
eps:
very small number setting the tolerance for determining equivalence between time values and for rounding non-integer values of deltat and frequency. Default : .Options$ts.eps
VALUE:
a multivariate time series.
DETAILS:
The multivariate time series is returned as a matrix or data frame with a time domain that is the union of the time domains of all the series input. NAs are used to represent time values at which there are no observations.

The resulting matrix has a dimnames attribute. If an argument was given a name in the function call, then the corresponding column has that name; if the argument was itself a matrix, then its name is expanded with digits denoting its respective columns. If the argument is a matrix which has a dimnames attribute, then these are used for the corresponding column names.

You cannot mix calendar and non-calendar time domains.

Unlike ts.union, the function cbind, which allows time series, vectors or matrices as arguments, does not return a time series and makes no effort to make time parameters consistent.

SEE ALSO:
ts.intersect
EXAMPLES: 
x <- rts( rnorm(300), start = 1, freq = 12)
y <- rts( rnorm(100), start = 1, freq = 4)
noisy <- ts.union(x,y)