Compute Lee's statistic
A simple function to compute Lee's L statistic for bivariate spatial data;
L(x,y) = (n sum_i (sum_j w_ij (x_i - xbar)) (sum_j w_ij (y_j - ybar))) / (S2 sqrt(sum_i (x_i - xbar)^2)) sqrt(sum_i (x_i - xbar)^2))
lee(x, y, listw, n, S2, zero.policy=NULL, NAOK=FALSE)
x |
a numeric vector the same length as the neighbours list in listw |
y |
a numeric vector the same length as the neighbours list in listw |
listw |
a |
n |
number of zones |
S2 |
Sum of squared sum of weights by rows. |
zero.policy |
default NULL, use global option value; if TRUE assign zero to the lagged value of zones without neighbours, if FALSE assign NA |
NAOK |
if 'TRUE' then any 'NA' or 'NaN' or 'Inf' values in x are passed on to the foreign function. If 'FALSE', the presence of 'NA' or 'NaN' or 'Inf' values is regarded as an error. |
a list of
L |
Lee's L statistic |
local L |
Lee's local L statistic |
Roger Bivand and Virgiio Gómez-Rubio Virgilio.Gomez@uclm.es
Lee (2001). Developing a bivariate spatial association measure: An integration of Pearson's r and Moran's I. J Geograph Syst 3: 369-385
data(boston, package="spData") lw<-nb2listw(boston.soi) x<-boston.c$CMEDV y<-boston.c$CRIM z<-boston.c$RAD Lxy<-lee(x, y, lw, length(x), zero.policy=TRUE) Lxz<-lee(x, z, lw, length(x), zero.policy=TRUE)
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.