spatstat (version 1.55-1)

model.images: Compute Images of Constructed Covariates

Description

For a point process model fitted to spatial point pattern data, this function computes pixel images of the covariates in the design matrix.

Usage

model.images(object, ...)

# S3 method for ppm model.images(object, W = as.owin(object), ...)

# S3 method for kppm model.images(object, W = as.owin(object), ...)

# S3 method for dppm model.images(object, W = as.owin(object), ...)

# S3 method for lppm model.images(object, L = as.linnet(object), ...)

# S3 method for slrm model.images(object, ...)

Arguments

object

The fitted point process model. An object of class "ppm" or "kppm" or "lppm" or "slrm" or "dppm".

W

A window (object of class "owin") in which the images should be computed. Defaults to the window in which the model was fitted.

L

A linear network (object of class "linnet") in which the images should be computed. Defaults to the network in which the model was fitted.

Other arguments (such as na.action) passed to model.matrix.lm.

Value

A list (of class "solist") or array (of class "hyperframe") containing pixel images (objects of class "im"). For model.images.lppm, the images are also of class "linim".

Details

This command is similar to model.matrix.ppm except that it computes pixel images of the covariates, instead of computing the covariate values at certain points only.

The object must be a fitted spatial point process model object of class "ppm" (produced by the model-fitting function ppm) or class "kppm" (produced by the fitting function kppm) or class "dppm" (produced by the fitting function dppm) or class "lppm" (produced by lppm) or class "slrm" (produced by slrm).

The spatial covariates required by the model-fitting procedure are computed at every pixel location in the window W. For lppm objects, the covariates are computed at every location on the network L. For slrm objects, the covariates are computed on the pixels that were used to fit the model.

Note that the spatial covariates computed here are not the original covariates that were supplied when fitting the model. Rather, they are the covariates that actually appear in the loglinear representation of the (conditional) intensity and in the columns of the design matrix. For example, they might include dummy or indicator variables for different levels of a factor, depending on the contrasts that are in force.

The pixel resolution is determined by W if W is a mask (that is W$type = "mask"). Otherwise, the pixel resolution is determined by spatstat.options.

The format of the result depends on whether the original point pattern data were marked or unmarked.

  • If the original dataset was unmarked, the result is a named list of pixel images (objects of class "im") containing the values of the spatial covariates. The names of the list elements are the names of the covariates determined by model.matrix.lm. The result is also of class "solist" so that it can be plotted immediately.

  • If the original dataset was a multitype point pattern, the result is a hyperframe with one column for each possible type of points. Each column is a named list of pixel images (objects of class "im") containing the values of the spatial covariates. The row names of the hyperframe are the names of the covariates determined by model.matrix.lm.

See Also

model.matrix.ppm, model.matrix, ppm, ppm.object, lppm, dppm, kppm, slrm, im, im.object, plot.solist, spatstat.options

Examples

Run this code
# NOT RUN {
   fit <- ppm(cells ~ x)
   model.images(fit)
   B <- owin(c(0.2, 0.4), c(0.3, 0.8))
   model.images(fit, B)
   fit2 <- ppm(cells ~ cut(x,3))
   model.images(fit2)
   fit3 <- slrm(japanesepines ~ x)
   model.images(fit3)
   fit4 <- ppm(amacrine ~ marks + x)
   model.images(fit4)
# }

Run the code above in your browser using DataCamp Workspace