est_intensity_np: Estimate the intensity surface using a non-parametric method

In sppmix: Modeling Spatial Poisson and Related Point Processes

Description

Using an Epanechnikov kernel we calculate an estimate of intensity surface while accounting for edge effects.

For examples see

http://faculty.missouri.edu/~micheasa/sppmix/sppmix_all_examples.html#est_intensity_np

Usage

est_intensity_np(pattern, win, h, L = 10, kernel = c("Epanechnikov"),
  edgecorrect = TRUE, truncate = TRUE)

Arguments

pattern	A two-dimesional spatial point pattern in the form of a ppp object.
win	Object of class owin.
h	Kernel bandwidth. h should be a positive number.
L	Length of the side of the square grid.
kernel	Kernel used to estimate the intensity surface. Currently, only supports the Epanechnikov kernel.
edgecorrect	Logical flag indicating whether or not to use edge-correction in the estimating intensity surface. The default is TRUE.
truncate	Logical flag indicating whether or not to use points only within the window. The default is TRUE.

Value

An object of class im.

Author(s)

Jiaxun Chen, Sakis Micheas

See Also

rnormmix, to_int_surf, rsppmix, plotmix_3d

Examples

mix1 <- rnormmix(5, sig0 = .01, df = 5, xlim=c(0, 5),ylim=c(0, 5))
intsurf1=to_int_surf(mix1,lambda = 40, win =spatstat::owin( c(0, 5),c(0, 5)))
plot(intsurf1)
pp1 <- rsppmix(intsurf1)
# estimate and plot the estimated intensity surface
surfNP1 <- est_intensity_np(pp1, win=spatstat::owin( c(0, 5),c(0, 5)), h=0.05,
 L=100,truncate = FALSE)
plotmix_3d(surfNP1,title1="Non parametric estimator of the intensity surface")
#truncate components to have all their mass in the window
surfNP2 <- est_intensity_np(pp1, win=spatstat::owin( c(0, 5),c(0, 5)), h=0.5, L=100)
plotmix_3d(surfNP2,title1="(Truncated) Non parametric estimator of the intensity surface")

sppmix documentation built on Jan. 13, 2021, 10:04 p.m.