localsummary: Summary plots of the fitted coefficient of a local spatio-temporal Poisson process or a local LGCP model

Description

The function breaks up the contribution of the local estimates to the fitted intensity, by plotting the overall intensity and the density kernel smoothing of some artificial intensities, obtained by imputing the quartiles of the local parameters' distributions.

Usage

localsummary(
  x,
  scaler = c("silverman", "IQR", "sd", "var"),
  do.points = TRUE,
  print.bw = FALSE,
  zap = 1e-05,
  par = TRUE
)

Arguments

x: An object of class locstppm or stlgcppm
scaler: Optional. Controls the value for a scalar representation of the spatial scale of the data. Either a character string, "silverman" (default), "IQR", "sd", or "var"; or positive numeric value(s). See OS.
do.points: Add points to plot
print.bw: It prints the estimated oversmoothing (OS) bandwidth selector
zap: Noise threshold factor (default to 0.00001). A numerical value greater than or equal to 1. If the range of pixel values is less than zap * .Machine$double.eps, the image will be treated as constant. This avoids displaying images which should be constant but contain small numerical errors.
par: Default to TRUE.

Author

Nicoletta D'Angelo and Giada Adelfio

References

D'Angelo, N., Adelfio, G., and Mateu, J. (2023). Locally weighted minimum contrast estimation for spatio-temporal log-Gaussian Cox processes. Computational Statistics & Data Analysis, 180, 107679.

Davies, T.M. and Hazelton, M.L. (2010). Adaptive kernel estimation of spatial relative risk, Statistics in Medicine, 29(23) 2423-2437.

Terrell, G.R. (1990). The maximal smoothing principle in density estimation, Journal of the American Statistical Association, 85, 470-477.

Examples

Run this code

# \donttest{
# Local spatio-temporal Poisson process model

set.seed(2)
inh <- rstpp(lambda = function(x, y, t, a) {exp(a[1] + a[2]*x)}, 
             par = c(0.005, 5))
inh_local <- locstppm(inh, formula = ~ x)

localsummary(inh_local)

# Local LGCP

catsub <- stp(greececatalog$df[1:200, ])

lgcp_loc <- stlgcppm(catsub, formula = ~ x, first = "local")

localsummary(lgcp_loc)
# }