cvloglk {spatialkernel}R Documentation

Cross-Validated Log-Likelihood Function

Description

Calculate the cross-validated log-likelihood function.

Usage

  cvloglk(pts, marks, t = NULL, h)

Arguments

pts matrix containing the x,y-coordinates of the point locations.
marks numeric/character vector of the marked labels of the type of each point.
t numeric vector of the associated time-periods, default NULL for pure spatial data.
h numeric vector of the kernel smoothing bandwidths at which to calculate the cross-validated log-likelihood function.

Details

Select a common bandwidth for kernel regression estimation of type-specific probabilities of a multivariate Poisson point process with independent component processes of each categorical type by maximizing the cross-validate log-likelihood function.

Select a common bandwidth for kernel regression of type-specific probabilities for all time-periods when the argument t is not NULL, in which case the data is of a multivariate spatial-temporal point process, with t the values of associated time-periods.

Value

A list with components

cv vector of the values of the cross-validated Log-likelihood function.
hcv numeric value which maximizing the cross-validate log-likelihood function
... copy of the arguments pts, marks, h.

References

  1. Diggle, P.J., Zheng, P. and Durr, P. A. (2005) Nonparametric estimation of spatial segregation in a multivariate point process: bovine tuberculosis in Cornwall, UK. J. R. Stat. Soc. C, 54, 3, 645–658.

See Also

phat, mcseg.test, and mcpat.test


[Package spatialkernel version 0.4-8 Index]