| dlm {dlm} | R Documentation |
The function dlm is used to create Dynamic Linear Model objects.
as.dlm and is.dlm coerce an object to a Dynamic Linear
Model object and test whether an object is a Dynamic Linear Model.
dlm(...) as.dlm(obj) is.dlm(obj)
... |
A list with named elements m0,
C0, FF, V, GG, W and,
optionally,
JFF, JV, JGG, JW, and
X. The first six are the usual vector and matrices that define
a time-invariant DLM. The remaining elements are used for time-varying
DLM. X, if present, should be a matrix. If JFF is not
NULL, then it must be a matrix of the
same dimension of FF, with the (i,j) element being zero if
FF[i,j] is time-invariant, and a positive integer k
otherwise. In this case the (i,j) element of FF at time
t will be X[t,k]. A similar interpretation holds for
JV, JGG, and JW. ... may have additional
components, that are not used by dlm. The named components
may also be passed to the function as individual arguments. |
obj |
An arbitrary R object. |
The function dlm is used to create Dynamic Linear Model
objects. These are lists with the named elements described above and
with class of "dlm".
Class "dlm" has a number of methods. In particular, consistent
DLM can be added together to produce another DLM.
For dlm, an object of class "dlm".
Giovanni Petris GPetris@uark.edu
dlmModReg, dlmModPoly,
dlmModARMA, dlmModSeas, to create
particular objects of class "dlm".
## Linear regression as a DLM x <- matrix(rnorm(10),nc=2) mod <- dlmModReg(x) is.dlm(mod) ## Adding dlm's dlmModPoly() + dlmModSeas(4) # linear trend plus quarterly seasonal component