licd_multi function

Local Indicators for Categorical Data

Local indicators for categorical data combine a measure of local composition in a window given by the per-observation set of neighbouring observations, with a local multi-category joincount test simplified to neighbours with the same or different categories compared to the focal observation

licd_multi(fx, listw, zero.policy = attr(listw, "zero.policy"), adjust.n = TRUE,
 nsim = 0L, iseed = NULL, no_repeat_in_row = FALSE, control = list())

Arguments

• fx: a factor with two or more categories, of the same length as the neighbours and weights objects in listw
• listw: a listw object created for example by nb2listw
• zero.policy: default attr(listw, "zero.policy") as set when listw was created, if attribute not set, use global option value; if TRUE assign zero to the lagged value of zones without neighbours, if FALSE assign NA
• adjust.n: default TRUE, if FALSE the number of observations is not adjusted for no-neighbour observations, if TRUE, the number of observations is adjusted
• nsim: default 0, number of conditonal permutation simulations
• iseed: default NULL, used to set the seed; the output will only be reproducible if the count of CPU cores across which computation is distributed is the same
• no_repeat_in_row: default FALSE, if TRUE, sample conditionally in each row without replacements to avoid duplicate values, https://github.com/r-spatial/spdep/issues/124
• control: comp_binary=TRUE default TRUE, ignoring other weights styles than binary for composition measures, binomial_punif_alternative="greater", jcm_same_punif_alternative="less", jcm_diff_punif_alternative="greater", rank_ties.method="min" default "min", na.last="keep" leading to rank NA being returned if the observed joincount variance is non-positive; if TRUE joincount NAs are ranked highest when using rank, for others see ?rank, check_reps=FALSE, unique_ceiling=1/3 used if check_reps TRUE, check_reps=FALSE if TRUE, check and report how many unique draws are made among the nsim draws, and if the number of unique draws is less than unique_ceiling, compute measures only for unique draws and copy out to replicated draws, pysal_rank=FALSE use rank with rank_ties.method and na.last; if TRUE, use pysal-style ranking to find the rank of sum(sims <= obs, na.rm=TRUE)+1 for pysal_sim_obs "GE", sum(sims < obs, na.rm=TRUE)+1 for the count of observed greater than "GT" the simulated values, pysal_sim_obs="GT" may also be "GE", xtras=FALSE if TRUE return calculated compostion values of BW chi-squared, k-colour chi-squared, and BW Anscombe

Details

The original code may be found at tools:::Rd_expr_doi("10.5281/zenodo.4283766")

Returns

• local_comp: data.frame object with LICD local composition columns: ID, category_i, count_like_i, prop_i, count_nbs_i, pbinom_like_BW, pbinom_unlike_BW, pbinom_unlike_BW_alt, chi_BW_i, chi_K_i, anscombe_BW

• local_config: data.frame object with LICD local configuration columns: ID, jcm_chi_obs, jcm_count_BB_obs, jcm_count_BW_obs, jcm_count_WW_obs, pval_jcm_obs_BB, pval_jcm_obs_WW, pval_jcm_obs_BW

• local_comp_sim: data.frame object with permutation-based LICD local composition columns: ID, pbinom_like_BW, pbinom_unlike_BW, pbinom_unlike_BW_alt, rank_sim_chi_BW, rank_sim_chi_K, rank_sim_anscombe

• local_config_sim: data.frame object with permutation-based LICD local configuration columns: ID, jcm_chi_sim_rank, pval_jcm_obs_BB, pval_jcm_obs_BW, pval_jcm_obs_WW

References

Cliff, A. D., Ord, J. K. 1981 Spatial processes, Pion, p. 20;

Upton, G., Fingleton, B. 1985 Spatial data analysis by example: point pattern and qualitative data, Wiley, pp. 158--170;

Boots, B., 2003. Developing local measures of spatial association for categorical data. Journal of Geographical Systems 5, 139–160;

Boots, B., 2006. Local configuration measures for categorical spatial data: binary regular lattices. Journal of Geographical Systems 8 (1), 1–24;

Pietrzak, M.B., Wilk, J., Kossowski, T., Bivand, R.S., 2014. The application of local indicators for categorical data (LICD) in the spatial analysis of economic development. Comparative Economic Research 17 (4), 203–220 tools:::Rd_expr_doi("10.2478/cer-2014-0041") ;

Bivand, R.S., Wilk, J., Kossowski, T., 2017. Spatial association of population pyramids across Europe: The application of symbolic data, cluster analysis and join-count tests. Spatial Statistics 21 (B), 339–361 tools:::Rd_expr_doi("10.1016/j.spasta.2017.03.003") ;

Francesco Carrer, Tomasz M. Kossowski, Justyna Wilk, Michał B. Pietrzak, Roger S. Bivand, The application of Local Indicators for Categorical Data (LICD) to explore spatial dependence in archaeological spaces, Journal of Archaeological Science, 126, 2021, tools:::Rd_expr_doi("10.1016/j.jas.2020.105306")

Author(s)

Roger Bivand Roger.Bivand@nhh.no based on earlier code by Tomasz M. Kossowski, Justyna Wilk and Michał B. Pietrzak

Note

In order to increase the numbers of neighbours using nblag and nblag_cumul is advisable; use of binary weights is advised and are in any case used for the composition measure

See Also

joincount.multi

Examples

columbus <- st_read(system.file("shapes/columbus.gpkg", package="spData")[1], quiet=TRUE)
HICRIME <- cut(columbus$CRIME, breaks=c(0,35,80), labels=c("low","high"))
(nb <- poly2nb(columbus))
lw <- nb2listw(nblag_cumul(nblag(nb, 2)), style="B")
obj <- licd_multi(HICRIME, lw)
str(obj)
h_obj <- hotspot(obj)
str(h_obj)
table(h_obj$both_recode)
columbus$both <- h_obj$both_recode
plot(columbus[, "both"])
GDAL37 <- as.numeric_version(unname(sf_extSoftVersion()["GDAL"])) >= "3.7.0"
file <- "etc/shapes/GB_2024_southcoast_50m.gpkg.zip"
zipfile <- system.file(file, package="spdep")
if (GDAL37) {
    sc50m <- st_read(zipfile)
} else {
    td <- tempdir()
    bn <- sub(".zip", "", basename(file), fixed=TRUE)
    target <- unzip(zipfile, files=bn, exdir=td)
    sc50m <- st_read(target)
}
sc50m$Winner <- factor(sc50m$Winner, levels=c("Con", "Green", "Lab", "LD"))
plot(sc50m[,"Winner"], pal=c("#2297E6", "#61D04F", "#DF536B", "#F5C710"))
nb_sc_50m <- poly2nb(sc50m, row.names=as.character(sc50m$Constituency))
sub2 <- attr(nb_sc_50m, "region.id")[attr(nb_sc_50m, "ncomp")$comp.id == 2L]
iowe <- match(sub2[1], attr(nb_sc_50m, "region.id"))
diowe <- c(st_distance(sc50m[iowe,], sc50m))
meet_criterion <- sum(diowe <= units::set_units(5000, "m"))
cands <- attr(nb_sc_50m, "region.id")[order(diowe)[1:meet_criterion]]
nb_sc_50m_iowe <- addlinks1(nb_sc_50m, from = cands[1],
 to = cands[3:meet_criterion])
ioww <- match(sub2[2], attr(nb_sc_50m, "region.id"))
dioww <- c(st_distance(sc50m[ioww,], sc50m))
meet_criterion <- sum(dioww <= units::set_units(5000, "m"))
cands <- attr(nb_sc_50m, "region.id")[order(dioww)[1:meet_criterion]]
nb_sc_50m_iow <- addlinks1(nb_sc_50m_iowe, from = cands[2], to = cands[3:meet_criterion])
nb_sc_1_2 <- nblag_cumul(nblag(nb_sc_50m_iow, 2))
lw <- nb2listw(nb_sc_1_2, style="B")
licd_obj <- licd_multi(sc50m$Winner, lw)
h_obj <- hotspot(licd_obj)
sc50m$both <- h_obj$both_recode
plot(sc50m[, "both"])
ljc <- local_joincount_uni(factor(sc50m$Winner == "LD"), chosen="TRUE", lw)
sc50m$LD_pv <- ljc[, 2]
plot(sc50m[, "LD_pv"], breaks=c(0, 0.025, 0.05, 0.1, 0.5, 1))