Export matrix weights to excel.

Yes, I got it.

Now, I saw an issue in the matrix, I?ve got two regions with value 1 
inside the matrix, if it is W style, how is it possible?. The rest of 
rows are standardized.
You can "see" it, nobody else can. Are all the rowSums() of the matrix 1? 
If so, two of the observations only have one neighbour (maybe a pair 
linked only to each other). Why would this be unexpected if you are using 
distance-based neighbours?
Another question, the output follow the order of regions of the shape 
file, isn?t?
They follow the order of the coordinate matrix you passed to dnearneigh(), 
but you can check this easily enough.

Please do read the documentation, none of this is particularly difficult.

Roger
Thank you
El 04/05/2014, a las 12:46, Roger Bivand <Roger.Bivand at nhh.no> escribi?:

On Sat, 3 May 2014, Rolando Valdez wrote:

Hello,

I'm working with a matrix weights, I want to know if I can export the output, to handle with excel.

This is a part what I have:

zm <- readOGR(".", "zmdis?)
OGR data source with driver: ESRI Shapefile
Source: ".", layer: "zmdis"
with 56 features and 8 fields
Feature type: wkbPolygon with 2 dimensions
mapaxy = coordinates(zm)
wdist100 = dnearneigh(mapaxy, d1=0, d2=100000)
l100 = nb2listw(wdist100, style="W", zero.policy=T)
summary(l100, zero.policy=T)
Characteristics of weights list object:
Neighbour list object:
Number of regions: 56
Number of nonzero links: 86
Percentage nonzero weights: 2.742347
Average number of links: 1.535714
21 regions with no links:
3 6 11 13 14 15 16 24 25 27 28 35 36 38 41 47 48 51 52 53 54
Link number distribution:

0  1  2  3  4  5  8
21 12 11  2  7  2  1
12 least connected regions:
1 5 7 17 19 20 21 30 33 34 45 55 with 1 link
1 most connected region:
37 with 8 links

Weights style: W
Weights constants summary:
 n   nn S0       S1    S2
W 35 1225 35 39.26667 147.3

I want to get a matrix [56 x 56] to work in excel with it.
?listw2mat should help, then write the matrix in a form that Excel can read. But note that your distance criterion leaves 21 of 56 observations with no neighbours.

Hope this helps,

Roger

Thank you in advance.

Rolando Valdez

_______________________________________________
R-sig-Geo mailing list
R-sig-Geo at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-geo

--
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; fax +47 55 95 91 00
e-mail: Roger.Bivand at nhh.no
Rolando Valdez

_______________________________________________
R-sig-Geo mailing list
R-sig-Geo at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-geo

Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; fax +47 55 95 91 00
e-mail: Roger.Bivand at nhh.no

Export matrix weights to excel.

Thread (4 messages)