bw.relrisk

0th

Percentile

Cross Validated Bandwidth Selection for Relative Risk Estimation

Uses cross-validation to select a smoothing bandwidth for the estimation of relative risk.

Keywords
methods, smooth, spatial
Usage
bw.relrisk(X, method = "likelihood", nh = 32)
Arguments
X
A multitype point pattern (object of class "ppp" which has factor valued marks).
method
Character string determining the cross-validation method. Current options are "likelihood", "leastsquares" or "weightedleastsquares".
nh
Number of trial values of smoothing bandwith sigma to consider.
Details

This function selects an appropriate bandwidth for the nonparametric estimation of relative risk using relrisk. Consider the indicators $y_{ij}$ which equal $1$ when data point $x_i$ belongs to type $j$, and equal $0$ otherwise. For a particular value of smoothing bandwidth, let $\hat p_j(u)$ be the estimated probabilities that a point at location $u$ will belong to type $j$. Then the bandwidth is chosen to minimise either the likelihood, the squared error, or the approximately standardised squared error, of the indicators $y_{ij}$ relative to the fitted values $\hat p_j(x_i)$. See Diggle (2003).

The result is a numerical value giving the selected bandwidth sigma. The result also belongs to the class "bw.relrisk" allowing it to be printed and plotted. The plot shows the cross-validation criterion as a function of bandwidth.

Value

  • A numerical value giving the selected bandwidth. The result also belongs to the class "bw.relrisk".

References

Diggle, P.J. (2003) Statistical analysis of spatial point patterns, Second edition. Arnold.

See Also

relrisk

Aliases
  • bw.relrisk
Examples
data(urkiola)
  b <- bw.relrisk(urkiola)
  b
  plot(b)
Documentation reproduced from package spatstat, version 1.23-3, License: GPL (>= 2)

Community examples

Looks like there are no examples yet.