hotspot: Cluster Classifications for Local Indicators of Spatial Association and Local Indicators for Categorical Data

Description

Used to return a factor showing so-called cluster classification for local indicators of spatial association for local Moran's I, local Geary's C (and its multivariate variant) and local Getis-Ord G. This factor vector can be added to a spatial object for mapping. When obj is of class licd, a list of up to six factors for measures of local composition (analytical and permutation), local configuration (analytical and permutation), and combined measures, both the interaction of composition and configuration, and a simplified recoding of these.

Usage

hotspot(obj, ...)
# S3 method for default
hotspot(obj, ...)
# S3 method for localmoran
hotspot(obj, Prname, cutoff=0.005, quadrant.type="mean",
 p.adjust="fdr", droplevels=TRUE, ...)
# S3 method for summary.localmoransad
hotspot(obj, Prname, cutoff=0.005,
 quadrant.type="mean", p.adjust="fdr", droplevels=TRUE, ...)
# S3 method for data.frame.localmoranex
hotspot(obj, Prname, cutoff=0.005,
 quadrant.type="mean", p.adjust="fdr", droplevels=TRUE, ...)
# S3 method for localG
hotspot(obj, Prname, cutoff=0.005, p.adjust="fdr", droplevels=TRUE, ...)
# S3 method for localC
hotspot(obj, Prname, cutoff=0.005, p.adjust="fdr", droplevels=TRUE, ...)
# S3 method for licd
hotspot(obj, type = "both", cutoff = 0.05, p.adjust = "none", 
 droplevels = TRUE, control = list(), ...)

Value

A factor showing so-called cluster classification for local indicators of spatial association. When obj is of class licd, a list of up to six factors for measures of local composition (analytical and permutation), local configuration (analytical and permutation), and combined measures, both the interaction of composition and configuration, and a simplified recoding of these.

Arguments

obj: An object of class localmoran, localC or localG
Prname: A character string, the name of the column containing the probability values to be classified by cluster type if found “interesting”
cutoff: Default 0.005, the probability value cutoff larger than which the observation is not found “interesting”
p.adjust: Default "fdr", the p.adjust() method used, one of c("holm", "hochberg", "hommel", "bonferroni", "BH", "BY", "fdr", "none")
droplevels: Default TRUE, should empty levels of the input cluster factor be dropped
quadrant.type: Default "mean", for "localmoran" objects only, can be c("mean", "median", "pysal") to partition the Moran scatterplot; "mean" partitions on the means of the variable and its spatial lag, "median" on medians of the variable and its spatial lag, "pysal" at zero for the centred variable and its spatial lag
type: When obj is of class licd, default both, may also be comp for local composition or config for local configuration
control: When obj is of class licd, default binomial_sidak 2, binomial_overlap TRUE, jcm_sidak 3. binomial_overlap may be set FALSE to avoid the Binomial probability values summing to more than unity - the tests in Boots (2003, p. 141) do overlap (>= and <=), and the Šidák exponents may be set to 1 to prevent by-observation correction for 2 Binomial and 3 Normal probability values per observation
...: other arguments passed to methods.

Author

Roger Bivand

Examples

Run this code

orig <- spData::africa.rook.nb
listw <- nb2listw(orig)
x <- spData::afcon$totcon

set.seed(1)
C <- localC_perm(x, listw)
Ch <- hotspot(C, Prname="Pr(z != E(Ci)) Sim", cutoff=0.05, p.adjust="none")
table(addNA(Ch))
set.seed(1)
I <- localmoran_perm(x, listw)
Ih <- hotspot(I, Prname="Pr(z != E(Ii)) Sim", cutoff=0.05, p.adjust="none")
table(addNA(Ih))
Is <- summary(localmoran.sad(lm(x ~ 1), nb=orig))
Ish <- hotspot(Is, Prname="Pr. (Sad)", cutoff=0.05, p.adjust="none")
table(addNA(Ish))
Ie <- as.data.frame(localmoran.exact(lm(x ~ 1), nb=orig))
Ieh <- hotspot(Ie, Prname="Pr. (exact)", cutoff=0.05, p.adjust="none")
table(addNA(Ieh))
set.seed(1)
G <- localG_perm(x, listw)
Gh <- hotspot(G, Prname="Pr(z != E(Gi)) Sim", cutoff=0.05, p.adjust="none")
table(addNA(Gh))