spdep (version 1.3-7)

local_joincount_bv: Calculate the local bivariate join count

Description

The bivariate join count (BJC) evaluates event occurrences in predefined regions and tests if the co-occurrence of events deviates from complete spatial randomness.

Usage

local_joincount_bv(
  x,
  z,
  listw,
  nsim = 199,
  alternative = "two.sided"
)

Value

a data.frame with two columns join_count and p_sim and number of rows equal to the length of arguments x.

Arguments

x

a binary variable either numeric or logical

z

a binary variable either numeric or logical with the same length as x

listw

a listw object containing binary weights created, for example, with nb2listw(nb, style = "B")

nsim

the number of conditional permutation simulations

alternative

default "greater". One of "less" or "greater".

Author

Josiah Parry

Details

There are two cases that are evaluated in the bivariate join count. The first being in-situ colocation (CLC) where xi = 1 and zi = 1. The second is the general form of the bivariate join count (BJC) that is used when there is no in-situ colocation.

The BJC case "is useful when x and z cannot occur in the same location, such as when x and z correspond to two different values of a single categorical variable" or "when x and z can co-locate, but do not" (Anselin and Li, 2019). Whereas the CLC case is useful in evaluating simultaneous occurrences of events.

The local bivariate join count statistic requires a binary weights list which can be generated with nb2listw(nb, style = "B").

P-values are only reported for those regions that match the CLC or BJC criteria. Others will not have an associated p-value.

P-values are estimated using a conditional permutation approach. This creates a reference distribution from which the observed statistic is compared.

References

Anselin, L., & Li, X. (2019). Operational Local Join Count Statistics for Cluster Detection. Journal of geographical systems, 21(2), 189–210. tools:::Rd_expr_doi("10.1007/s10109-019-00299-x")

Examples

data("oldcol")
listw <- nb2listw(COL.nb, style = "B")
# Colocation case
x <- COL.OLD[["CP"]]
z <- COL.OLD[["EW"]]
set.seed(1)
res <- local_joincount_bv(x, z, listw)
na.omit(res)
# no colocation case
z <- 1 - x
set.seed(1)
res <- local_joincount_bv(x, z, listw)
na.omit(res)