Dear all,
I'm using the "skater" function in spdep in order to perform a spatial
clustering of polygons containing fire regime data. However, I have the
following problem. When I compute the cost, some of my polygons are similar to
all their neighbors; hence, all edges have 0 cost values for these polygons.
When I compute the nb2listw using the neighbors list and the glist as cost, it
obviously reports that :
>In nb2listw(ecodist_80_99.nb, lcosts_ecodist_80_99, style = "W", :
zero sum general weights
For those who are familiar with spatial clustering, I'm looking for
proposition. Maybe the only issue would be to dissolve these polygons and then
rerun the analyses. Morever, is there another spatial clustering technique in R
that can cope with such data?
Thanks in advance,
Yan
Yan Boulanger
Dendrolab - Datation de bâtiments patrimoniaux par dendrochronologie
www.dendrolab.ca
i...@dendrolab.ca
Université du Québec à Rimouski
Département de Biologie
Centre d'études nordiques
300, allée des Ursulines
Rimouski, PQ G5L 3A1
(418) 723-1986 poste 1482
[[alternative HTML version deleted]]
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo@stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
