Compute Moran's I
A simple function to compute Moran's I, called by moran.test and moran.mc;
I = (n sum_i sum_j w_ij (x_i - xbar) (x_j - xbar)) / (S0 sum_i (x_i - xbar)^2)
moran(x, listw, n, S0, zero.policy=NULL, NAOK=FALSE)
x |
a numeric vector the same length as the neighbours list in listw |
listw |
a |
n |
number of zones |
S0 |
global sum of weights |
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
I |
Moran's I |
K |
sample kurtosis of x |
Roger Bivand Roger.Bivand@nhh.no
Cliff, A. D., Ord, J. K. 1981 Spatial processes, Pion, p. 17.
data(oldcol) col.W <- nb2listw(COL.nb, style="W") crime <- COL.OLD$CRIME str(moran(crime, col.W, length(COL.nb), Szero(col.W))) is.na(crime) <- sample(1:length(crime), 10) str(moran(crime, col.W, length(COL.nb), Szero(col.W), NAOK=TRUE))
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.