RMmodel
RFboxcox
performs the Box-Cox transformation:
$\frac{(x+\mu)^\lambda-1}{\lambda}$
RFboxcox(data, boxcox, vdim = 1, inverse=FALSE, ignore.na=FALSE)
boxcox
should be a $2 \times
m$ matrix.
FALSE
an error message is returned
if any value of boxcox
is NA
. Otherwise the data are
returned without being transformed.
RFboxcox
returns a list
of three components, Y
, X
, vdim
returning
the deterministic trend, the design matrix, and the multivariability,
respectively.
If set
is positive, Y
and X
contain
the values for the set
-th set of coordinates.
Else, Y
and X
are both lists containing
the values for all the sets.
boxcox
can be set
globally through RFoptions
. If it is set globally the
transformation applies in the Gaussian case to
RFfit
,
RFsimulate
,
RFinterpolate
,
RFempiricalvariogram
.
Always first, the Box-Cox transformation is applied to the data.
Then the command is performed. The result is back-transformed before
returned.
If the first value of the transformation is Inf
no
transformation is performed (and is identical to boxcox = c(1,
0)
).
If boxcox
has length 1, then the transformation parameters
$\mu$ is set to $0$, which is the standard case.
RMmodel
,
RFsimulate
,
RFlikelihood
.
RFoptions(seed=0) ## *ANY* simulation will have the random seed 0; set
## RFoptions(seed=NA) to make them all random again
data(soil)
str(soil)
soil <- RFspatialPointsDataFrame(
coords = soil[ , c("x.coord", "y.coord")],
data = soil[ , c("moisture", "NO3.N", "Total.N", "NH4.N", "DOC", "N20N")],
RFparams=list(vdim=6, n=1)
)
data <- soil["moisture"]
model <- ~1 + RMplus(RMwhittle(scale=NA, var=NA, nu=NA), RMnugget(var=NA))
## Assuming log-Gaussian Data
print(fit <- RFfit(model, data=data, loggaus=TRUE))
## main Parameter in the Box Cox transformation to be estimated
print(fit <- RFfit(model, data=data, boxcox=NA))
Run the code above in your browser using DataLab