choynowski: Choynowski probability map values

In spdep: Spatial Dependence: Weighting Schemes, Statistics and Models

Description

Calculates Choynowski probability map values.

Usage

choynowski(n, x, row.names=NULL, tol = .Machine$double.eps^0.5, legacy=FALSE)

Arguments

n	a numeric vector of counts of cases
x	a numeric vector of populations at risk
row.names	row names passed through to output data frame
tol	accumulate values for observed counts >= expected until value less than tol
legacy	default FALSE using vectorised alternating side ppois version, if true use original version written from sources and iterating down to tol

Value

A data frame with columns:

pmap	Poisson probability map values: probablility of getting a more “extreme” count than actually observed, one-tailed with less than expected and more than expected folded together
type	logical: TRUE if observed count less than expected

Author(s)

Roger Bivand [e-mail]

References

Choynowski, M (1959) Maps based on probabilities, Journal of the American Statistical Association, 54, 385–388; Cressie, N, Read, TRC (1985), Do sudden infant deaths come in clusters? Statistics and Decisions, Supplement Issue 2, 333–349; Bailey T, Gatrell A (1995) Interactive Spatial Data Analysis, Harlow: Longman, pp. 300–303.

See Also

probmap

Examples

example(auckland)
res <- choynowski(auckland$M77_85, 9*auckland$Und5_81)
resl <- choynowski(auckland$M77_85, 9*auckland$Und5_81, legacy=TRUE)
all.equal(res, resl)
rt <- sum(auckland$M77_85)/sum(9*auckland$Und5_81)
ch_ppois_pmap <- numeric(length(auckland$Und5_81))
side <- c("greater", "less")
for (i in seq(along=ch_ppois_pmap)) {
  ch_ppois_pmap[i] <- poisson.test(auckland$M77_85[i], r=rt,
    T=(9*auckland$Und5_81[i]), alternative=side[(res$type[i]+1)])$p.value
}
all.equal(ch_ppois_pmap, res$pmap)

res1 <- probmap(auckland$M77_85, 9*auckland$Und5_81)
table(abs(res$pmap - res1$pmap) < 0.00001, res$type)
lt005 <- (res$pmap < 0.05) & (res$type)
ge005 <- (res$pmap < 0.05) & (!res$type)
cols <- rep("white", length(lt005))
cols[lt005] <- grey(2/7)
cols[ge005] <- grey(5/7)
plot(auckland, col=cols) 
legend("bottomleft", fill=grey(c(2,5)/7), legend=c("low", "high"), bty="n")

Example output

Loading required package: sp
Loading required package: Matrix

acklnd> require(maptools)
Loading required package: maptools
Checking rgeos availability: TRUE

acklnd> auckland <- readShapePoly(system.file("etc/shapes/auckland.shp",
acklnd+  package="spdep")[1])

acklnd> auckland.nb <- poly2nb(auckland)
Warning message:
use rgdal::readOGR or sf::st_read 
[1] TRUE
[1] TRUE
       
        FALSE TRUE
  FALSE    74    0
  TRUE      0   93

spdep documentation built on Aug. 19, 2017, 3:01 a.m.