Performs spatial smoothing of numeric values observed at a set of irregular locations using inverse-distance weighting.
idw(X, power=2, at=c("pixels", "points"), ..., se=FALSE)If X has a single column of marks:
If at="pixels" (the default), the result is
a pixel image (object of class "im").
Pixel values are values of the interpolated function.
If at="points", the result is a numeric vector
of length equal to the number of points in X.
Entries are values of the interpolated function at the points of X.
If X has a data frame of marks:
If at="pixels" (the default), the result is a named list of
pixel images (object of class "im"). There is one
image for each column of marks. This list also belongs to
the class "solist", for which there is a plot method.
If at="points", the result is a data frame
with one row for each point of X,
and one column for each column of marks.
Entries are values of the interpolated function at the points of X.
If se=TRUE, then the result is a list
with two entries named estimate and SE, which
each have the format described above.
A marked point pattern (object of class "ppp").
Numeric. Power of distance used in the weighting.
Character string specifying whether to compute the intensity values
at a grid of pixel locations (at="pixels") or
only at the points of X (at="points").
String is partially matched.
Arguments passed to as.mask
to control the pixel resolution of the result.
Logical value specifying whether to calculate a standard error.
Adrian Baddeley Adrian.Baddeley@curtin.edu.au, Rolf Turner rolfturner@posteo.net and Ege Rubak rubak@math.aau.dk. Variance calculation by Andrew P Wheeler with modifications by Adrian Baddeley.
This function performs spatial smoothing of numeric values observed at a set of irregular locations.
Smoothing is performed by inverse distance weighting. If the observed values are \(v_1,\ldots,v_n\) at locations \(x_1,\ldots,x_n\) respectively, then the smoothed value at a location \(u\) is $$ g(u) = \frac{\sum_i w_i v_i}{\sum_i w_i} $$ where the weights are the inverse \(p\)-th powers of distance, $$ w_i = \frac 1 {d(u,x_i)^p} $$ where \(d(u,x_i) = ||u - x_i||\) is the Euclidean distance from \(u\) to \(x_i\).
The argument X must be a marked point pattern (object
of class "ppp", see ppp.object).
The points of the pattern are taken to be the
observation locations \(x_i\), and the marks of the pattern
are taken to be the numeric values \(v_i\) observed at these
locations.
The marks are allowed to be a data frame. Then the smoothing procedure is applied to each column of marks.
If at="pixels" (the default), the smoothed mark value
is calculated at a grid of pixels, and the result is a pixel image.
The arguments ... control the pixel resolution.
See as.mask.
If at="points", the smoothed mark values are calculated
at the data points only, using a leave-one-out rule (the mark value
at a data point is excluded when calculating the smoothed value
for that point).
An estimate of standard error is also calculated, if se=TRUE.
The calculation assumes that the data point locations are fixed,
that is, the standard error only takes into account the variability
in the mark values, and not the variability due to randomness of the
data point locations.
An alternative to inverse-distance weighting is kernel smoothing,
which is performed by Smooth.ppp.
Shepard, D. (1968) A two-dimensional interpolation function for irregularly-spaced data. Proceedings of the 1968 ACM National Conference, 1968, pages 517--524. DOI: 10.1145/800186.810616
density.ppp,
ppp.object,
im.object.
See Smooth.ppp for kernel smoothing
and nnmark for nearest-neighbour interpolation.
To perform other kinds of interpolation, see also the akima package.
# data frame of marks: trees marked by diameter and height
plot(idw(finpines))
idw(finpines, at="points")[1:5,]
plot(idw(finpines, se=TRUE)$SE)
idw(finpines, at="points", se=TRUE)$SE[1:5, ]
Run the code above in your browser using DataLab