seg.ind: Dixon's Segregation Indices for NNCTs

In elvanceyhan/nnspat: Nearest Neighbor Methods for Spatial Patterns

Dixon's Segregation Indices for NNCTs

Description

Returns Dixon's segregation indices in matrix form based on entries of the NNCT, ct. Segregation index for cell i,j is defined as log(N_{ii}(n-n_i)/((n_i-N_{ii})(n_i-1)) if i=j and as log(N_{ij}(n-n_j-1)/((n_i-N_{ij})(n_j)) if i \ne j. See (\insertCitedixon:NNCTEco2002,ceyhan:SiM-seg-ind2014;textualnnspat).

The argument inf.corr is a logical argument (default=FALSE) to avoid \pm ∞ for the segregation indices. If TRUE indices are modified so that they are finite and if FALSE the above definition is used. (See \insertCiteceyhan:SiM-seg-ind2014;textualnnspat for more detail).

Usage

seg.ind(ct, inf.corr = FALSE)

Arguments

ct	A contingency table, in particular an NNCT
inf.corr	A logical argument (default=FALSE). If TRUE, indices are modified so that they are finite and if FALSE the above definition in the description is used.

Value

Returns a matrix of segregation indices which is of the same dimension as ct.

Author(s)

Elvan Ceyhan

See Also

Pseg.coeff, seg.coeff, Zseg.ind and Zseg.ind.ct

Examples

n<-20  #or try sample(1:20,1)
Y<-matrix(runif(3*n),ncol=3)
ipd<-ipd.mat(Y)
cls<-sample(1:2,n,replace = TRUE)  #or try cls<-rep(1:2,c(10,10))
ct<-nnct(ipd,cls)
ct
seg.ind(ct)
seg.ind(ct,inf.corr = TRUE)

#cls as a factor
na<-floor(n/2); nb<-n-na
fcls<-rep(c("a","b"),c(na,nb))
ct<-nnct(ipd,fcls)

seg.ind(ct)

#############
n<-40
Y<-matrix(runif(3*n),ncol=3)
ipd<-ipd.mat(Y)
cls<-sample(1:4,n,replace = TRUE)  #or try cls<-rep(1:2,c(10,10))
ct<-nnct(ipd,cls)

seg.ind(ct)
seg.ind(ct,inf.corr = TRUE)

ct<-matrix(c(0,10,5,5),ncol=2)
seg.ind(ct)

seg.ind(ct,inf.corr = TRUE)

elvanceyhan/nnspat documentation built on Sept. 4, 2022, 10:28 a.m.