spdep (version 0.8-1)

sp.correlogram: Spatial correlogram

Description

Spatial correlograms for Moran's I and the autocorrelation coefficient, with print and plot helper functions.

Usage

sp.correlogram(neighbours, var, order = 1, method = "corr",
 style = "W", randomisation = TRUE, zero.policy = NULL, spChk=NULL)
# S3 method for spcor
plot(x, main, ylab, ylim, ...)
# S3 method for spcor
print(x, p.adj.method="none", ...)

Arguments

neighbours

an object of class nb

var

a numeric vector

order

maximum lag order

method

"corr" for correlation, "I" for Moran's I, "C" for Geary's C

style

style can take values W, B, C, and S

randomisation

variance of I or C calculated under the assumption of randomisation, if FALSE normality

zero.policy

default NULL, use global option value; if FALSE stop with error for any empty neighbour sets, if TRUE permit the weights list to be formed with zero-length weights vectors

spChk

should the data vector names be checked against the spatial objects for identity integrity, TRUE, or FALSE, default NULL to use get.spChkOption()

x

an object from sp.correlogram() of class spcor

p.adj.method

correction method as in p.adjust

main

an overall title for the plot

ylab

a title for the y axis

ylim

the y limits of the plot

…

further arguments passed through

Value

returns a list of class spcor:

res

for "corr" a vector of values; for "I", a matrix of estimates of "I", expectations, and variances

method

"I" or "corr"

cardnos

list of tables of neighbour cardinalities for the lag orders used

var

variable name

Details

The print function also calculates the standard deviates of Moran's I or Geary's C and a two-sided probability value, optionally using p.adjust to correct by the nymber of lags. The plot function plots a bar from the estimated Moran's I, or Geary's C value to +/- twice the square root of its variance (in previous releases only once, not twice). The table includes the count of included observations in brackets after the lag order. Care needs to be shown when interpreting results for few remaining included observations as lag order increases.

References

Cliff, A. D., Ord, J. K. 1981 Spatial processes, Pion, pp. 118--122, Martin, R. L., Oeppen, J. E. 1975 The identification of regional forecasting models using space-time correlation functions, Transactions of the Institute of British Geographers, 66, 95--118.

See Also

nblag, moran, p.adjust

Examples

# NOT RUN {
if (require(rgdal, quietly=TRUE)) {
example(nc.sids, package="spData")
ft.SID74 <- sqrt(1000)*(sqrt(nc.sids$SID74/nc.sids$BIR74) +
  sqrt((nc.sids$SID74+1)/nc.sids$BIR74))
tr.SIDS74 <- ft.SID74*sqrt(nc.sids$BIR74)
cspc <- sp.correlogram(ncCC89_nb, tr.SIDS74, order=8, method="corr",
 zero.policy=TRUE)
print(cspc)
plot(cspc)
Ispc <- sp.correlogram(ncCC89_nb, tr.SIDS74, order=8, method="I",
 zero.policy=TRUE)
print(Ispc)
print(Ispc, "bonferroni")
plot(Ispc)
Cspc <- sp.correlogram(ncCC89_nb, tr.SIDS74, order=8, method="C",
 zero.policy=TRUE)
print(Cspc)
print(Cspc, "bonferroni")
plot(Cspc)
drop.no.neighs <- !(1:length(ncCC89_nb) %in% which(card(ncCC89_nb) == 0))
sub.ncCC89.nb <- subset(ncCC89_nb, drop.no.neighs)
plot(sp.correlogram(sub.ncCC89.nb, subset(tr.SIDS74,  drop.no.neighs),
 order=8, method="corr"))
}
# }