eire {spdep}R Documentation

Eire data sets

Description

The eire.df data frame has 26 rows and 9 columns. In addition, polygons of the 26 counties are provided as a multipart polylist in eire.polys.utm (coordinates in km, projection UTM zone 30). Their centroids are in eire.coords.utm. The original Cliff and Ord binary contiguities are in eire.nb.

Usage

data(eire)

Format

This data frame contains the following columns:

A
Percentage of sample with blood group A
towns
Towns/unit area
pale
Beyond the Pale 0, within the Pale 1
size
number of blood type samples
ROADACC
arterial road network accessibility in 1961
OWNCONS
percentage in value terms of gross agricultural output of each county consumed by itself
POPCHG
1961 population as percentage of 1926
RETSALE
value of retail sales £000
INCOME
total personal income £000

Source

Upton and Fingleton 1985, - Bailey and Gatrell 1995, ch. 1 for blood group data, Cliff and Ord (1973), p. 107 for remaining variables (also after O'Sullivan, 1968). Polygon borders and Irish data sourced from Michael Tiefelsdorf's SPSS Saddlepoint bundle: http://geog-www.sbs.ohio-state.edu/faculty/tiefelsdorf/GeoStat.htm.

Examples

data(eire)
summary(eire.df$A)
brks <- round(fivenum(eire.df$A), digits=2)
cols <- rev(heat.colors(4))
library(maptools)
plot(eire.polys.utm,
 col=cols[findInterval(eire.df$A, brks)], forcefill=FALSE)
title(main="Percentage with blood group A in Eire")
legend(x=c(-50, 70), y=c(6120, 6050), leglabs(brks), fill=cols, bty="n")
plot(eire.polys.utm, forcefill=FALSE)
plot(eire.nb, eire.coords.utm, add=TRUE)
lA <- lag.listw(nb2listw(eire.nb), eire.df$A)
summary(lA)
moran.test(spNamedVec("A", eire.df), nb2listw(eire.nb))
geary.test(spNamedVec("A", eire.df), nb2listw(eire.nb))
cor(lA, eire.df$A)
moran.plot(spNamedVec("A", eire.df), nb2listw(eire.nb),
 labels=rownames(eire.df))
A.lm <- lm(A ~ towns + pale, data=eire.df)
summary(A.lm)
res <- residuals(A.lm)
brks <- c(min(res),-2,-1,0,1,2,max(res))
cols <- rev(cm.colors(6))
plot(eire.polys.utm, col=cols[findInterval(res, brks)], forcefill=FALSE)
title(main="Regression residuals")
legend(x=c(-50, 70), y=c(6120, 6050), legend=leglabs(brks), fill=cols,
  bty="n")
lm.morantest(A.lm, nb2listw(eire.nb))
lm.morantest.sad(A.lm, nb2listw(eire.nb))
lm.LMtests(A.lm, nb2listw(eire.nb), test="LMerr")
brks <- round(fivenum(eire.df$OWNCONS), digits=2)
cols <- grey(4:1/5)
plot(eire.polys.utm,
 col=cols[findInterval(eire.df$OWNCONS, brks)], forcefill=FALSE)
title(main="Percentage own consumption of agricultural produce")
legend(x=c(-50, 70), y=c(6120, 6050), legend=leglabs(brks),
  fill=cols, bty="n")
moran.plot(spNamedVec("OWNCONS", eire.df), nb2listw(eire.nb))
moran.test(spNamedVec("OWNCONS", eire.df), nb2listw(eire.nb))
e.lm <- lm(OWNCONS ~ ROADACC, data=eire.df)
res <- residuals(e.lm)
brks <- c(min(res),-2,-1,0,1,2,max(res))
cols <- rev(cm.colors(6))
plot(eire.polys.utm, col=cols[findInterval(res, brks)], forcefill=FALSE)
title(main="Regression residuals")
legend(x=c(-50, 70), y=c(6120, 6050), legend=leglabs(brks), fill=cm.colors(6),
  bty="n")
lm.morantest(e.lm, nb2listw(eire.nb))
lm.morantest.sad(e.lm, nb2listw(eire.nb))
lm.LMtests(e.lm, nb2listw(eire.nb), test="LMerr")
print(localmoran.sad(e.lm, eire.nb, select=1:nrow(eire.df)))

[Package spdep version 0.3-12 Index]