sp.runs.test: Compute the global spatial runs test.

In rominsal/spqdata: Testing for spatial structure of qualitative data in cross section

Description

This function compute the global spatial runs test for spatial independence of a categorical spatial data set.

Usage

sp.runs.test(formula = NULL, data = NULL, na.action, xf = NULL,
listw = listw, alternative = "two.sided" , regular = FALSE,
distr = "asymptotic", nsim = NULL,control = list())

Arguments

formula	a symbolic description of the factor (optional).
data	an (optional) data frame or a sf object containing the variable to testing for.
xf	a factor (optional).
listw	una lista de vecinos (tipo knn o nb) o una matrix W que indique el orden de cada $m_i-entorno$ (por ejemplo de inversa distancia). Para calcular el número de rachas en cada m_i-entorno debe establecerse un orden, por ejemplo del vecino más próximo al más lejano.
alternative	a character string specifying the alternative hypothesis, must be one of "two.sided" (default), "greater" or "less".
distr	distribution of the test "asymptotic" (default) or "bootstrap"
nsim	Number of permutations to obtain confidence intervals (CI). Default value is NULL to don't get CI of number of runs.
control	List of additional control arguments.

Details

El orden de las vecindades ($m_i-entornos$) es crítico.
Para obetener el número de rachas observadas en cada $m_i-entorno$, cada elemento debe asociarse a un conjunto de vecinos ordenados por proximidad. Tres clases de listas pueden incluirse para identificar $m_i-entornos$:

knn	Matrices de la clase knn que consideran los vecinos por orden de proximidad. Ver knn2knn_order
nb	Si los vecinos se obtiene a partir de un objeto sf, el código internamente llamará a la función nb2nb_order los ordenará en orden de proximidad de los centroides.
matrix	Si se introduce simplemente una matriz basasa en la inversa de la distancia, también se llamará internamente a la función nb2nb_order para transformar la matriz de la clase matrix a una matriz de la clase nb con vecinos ordenados.

Two alternative sets of arguments can be included in this function to compute the spatial runs test:

Alternative 1	A factor (xf) and a list of neighborhood (listw) of the class knn.
Alternative 2	A sf object (data) and formula to specify the factor. A list of neighborhood (listw)

Value

A object of the htest and sprunstest class

data.name	a character string giving the names of the data.
method	the type of test applied ().
SR	total number of runs
dnr	empirical distribution of the number of runs
statistic	Value of the homogeneity runs statistic. Negative sign indicates global homogeneity
alternative	a character string describing the alternative hypothesis.
p.value	p-value of the SRQ
pseudo.value	the pseudo p-value of the SRQ test if nsim is not NULL
MeanNeig	Mean of the Maximum number of neighborhood
MaxNeig	Maximum number of neighborhood
listw	The list of neighborhood
nsim	number of boots (only for boots version)
SRGP	nsim simulated values of statistic.
SRLP	matrix with the number of runs for eacl localization.

Control arguments

seedinit	Numerical value for the seed (only for boot version). Default value seedinit=123

Author(s)

Fernando López	[e-mail]
Román Mínguez	[e-mail]
Antonio Páez	[e-mail]
Manuel Ruiz	[e-mail]

@references

• Ruiz, M., López, F., and Páez, A. (2010). A test for global and local homogeneity of categorical data based on spatial runs. Geographical Analysis.

See Also

local.sp.runs.test, dgp.spq, Q.test,

Examples

# Case 1: SRQ test based on factor and knn
rm(list = ls())
n <- 100
cx <- runif(n)
cy <- runif(n)
x <- cbind(cx,cy)
listw <- knearneigh(cbind(cx,cy), k=3)
p <- c(1/6,3/6,2/6)
rho <- 0.5
xf <- dgp.spq(listw = listw, p = p, rho = rho)
srq <- sp.runs.test(xf = xf, listw = listw)
print(srq)
plot(srq)
# Version boots
control <- list(seedinit = 1255)
srq <- sp.runs.test(xf = xf, listw = listw, distr = "bootstrap" , nsim = 299, control = control)
print(srq)
plot(srq)

# Case 2: SRQ test with formula, a sf object (points) and knn
rm(list = ls())
data("FastFood")
x <- cbind(FastFood.sf$Lon,FastFood.sf$Lat)
listw <- spdep::knearneigh(x, k=4)
formula <- ~ Type
srq <- sp.runs.test(formula = formula, data = FastFood.sf, listw = listw)
print(srq)
plot(srq)
# Version boots
srq <- sp.runs.test(formula = formula, data = FastFood.sf, listw = listw, distr = "bootstrap", nsim = 199)
print(srq)
plot(srq)

# Case 3: SRQ test (permutation) using formula with a sf object (polygons) and nb
rm(list = ls())
library(sf)
fname <- system.file("shape/nc.shp", package="sf")
nc <- st_read(fname)
listw <- spdep::poly2nb(as(nc,"Spatial"), queen = FALSE)
p <- c(1/6,3/6,2/6)
rho = 0.5
co <- sf::st_coordinates(sf::st_centroid(nc))
nc$xf <- dgp.spq(listw = listw, p = p, rho = rho)
plot(nc["xf"])
formula <- ~ xf
srq <- sp.runs.test(formula = formula, data = nc, listw = listw, distr = "bootstrap", nsim = 399)
print(srq)
plot(srq)

# Case 4: SRQ test (Asymptotic) using formula with a sf object (polygons) and nb
rm(list = ls())
# PARA PROBAR ELEMENTOS SIN VECINOS
data(Spain)
listw <- spdep::poly2nb(spain.sf, queen = FALSE)
plot(spain.sf["Coast"])
formula <- ~ Coast
srq <- sp.runs.test(formula = formula, data = spain.sf, listw = listw)
print(srq)
plot(srq)
# Version boots
srq <- sp.runs.test(formula = formula, data = spain.sf, listw = listw, distr = "bootstrap", nsim = 299)
print(srq)
plot(srq)

# Case 5: SRQ test based on a distance matrix (inverse distance)
rm(list = ls())
N <- 100
cx <- runif(N)
cy <- runif(N)
data <- as.data.frame(cbind(cx,cy))
data <- st_as_sf(data,coords = c("cx","cy"))
n = dim(data)[1]
dis <- 1/matrix(as.numeric(st_distance(data,data)),ncol=n,nrow=n)
diag(dis) <- 0
dis <- (dis < quantile(dis,.10))*dis
p <- c(1/6,3/6,2/6)
rho <- 0.5
xf <- dgp.spq(listw = dis , p = p, rho = rho)
srq <- sp.runs.test(xf = xf, listw = dis)
print(srq)
plot(srq)

srq <- sp.runs.test(xf = xf, listw = dis, data = data)
print(srq)
plot(srq)
# Version boots
srq <- sp.runs.test(xf = xf, listw = dis, data = data, distr = "bootstrap", nsim = 299)
print(srq)
plot(srq)

# Case 6: SRQ test based on a distance matrix (inverse distance)
rm(list = ls())
data("FastFood")
n = dim(FastFood.sf)[1]
dis <- 1000000/matrix(as.numeric(st_distance(FastFood.sf,FastFood.sf)), ncol = n, nrow = n)
diag(dis) <- 0
dis <- (dis < quantile(dis,.005))*dis
p <- c(1/6,3/6,2/6)
rho = 0.5
co <- sf::st_coordinates(sf::st_centroid(FastFood.sf))
FastFood.sf$xf <- dgp.spq(p = p, listw = dis, rho = rho)
plot(FastFood.sf["xf"])
formula <- ~ xf
# Version boots
srq <- sp.runs.test(formula = formula, data = FastFood.sf, listw = dis, distr = "bootstrap", nsim = 299)
print(srq)
plot(srq)

rominsal/spqdata documentation built on Jan. 24, 2021, 11:11 a.m.