## Neighbour list object:## Number of regions: 49 ## Number of nonzero links: 196 ## Percentage nonzero weights: 8.163265 ## Average number of links: 4
Let us read the GPKG file with valid geometries in to ‘sf’ and ‘sp’
@@ -257,7 +260,7 @@
Contiguity neighbours for pol
eps<-sqrt(.Machine$double.eps)system.time(for(iin1:reps)NY8_sf_1_nb<-poly2nb(NY8_sf, queen=TRUE, snap=eps))/reps
## user system elapsed
-## 0.1662 0.0095 0.1764
+## 0.1832 0.0096 0.1937
Using spatial indices to check intersection of polygons is much
faster than the legacy method in poly2nb. From spdep
1.1-7, use is made of GEOS through sf to find candidate
@@ -326,32 +329,29 @@
However, in making the geometries valid, we change the geometries, so
the new sets of neighbours still differ from those made with the valid
geometries in the same ways as before imposing validity:
or indeed taking the centre of the largest inscribed circle (the
function returns a radius line segment, so we choose the central point,
not the point on the circle):
By default, the function uses dbscan::frNN() to build a
kd-tree in 2D or 3D which is then used to find distance neighbours. For
small n, the argument use_kd_tree=FALSE may speed up
computation a little by reverting to legacy code not building a kd-tree
first, but in general the differences are so small that the user will
not notice:
The WGS84 ellipsoid Great Circle distances differ a very little from
the s2 spherical distances, yielding output that here
diverges for two tract centroids:
Alternatively, spherical distances can be used with
dwithin=FALSE and s2::s2_closest_edges();
although running in similar time, s2::s2_closest_edges()
depends on the additional k= argument, which, if mis-set,
may miss valid neighbours:
Using s2::s2_dwithin_matrix() requires a second pass,
one for the lower bound, another for the upper bound, and a set
difference operation to find neighbours in the distance band:
We can examine the names of the columns of the data frame to see what
it contains — in fact some of the same columns that we will be examining
below, and some others which will be useful in cleaning the data
@@ -397,12 +399,11 @@
Printing the neighbour object shows that it is a neighbour list
object, with a very sparse structure — if displayed as a matrix, only
3.94% of cells would be filled. Objects of class
diff --git a/docs/articles/subgraphs.html b/docs/articles/subgraphs.html
new file mode 100644
index 00000000..893d76a1
--- /dev/null
+++ b/docs/articles/subgraphs.html
@@ -0,0 +1,702 @@
+
+
+
The spdep package has always been careful about
+disconnected graphs, especially where the disconnected observations are
+graph nodes with no neighbours, that is no incoming or outgoing edges.
+In nb neighbour objects, they are encoded as integer
+vectors of length 1 containing integer 0, which is an
+invalid index on
+,
+where
+
+is the observation count. Functions taking neighbour objects as
+arguments use the zero.policy argument to guide how to
+handle no-neighbour observations.
+
spdep has also had n.comp.nb to find the
+number of disjoint connected subgraphs in an nb object,
+contributed by Nicholas Lewin-Koh in 2001 and using depth-first search
+for symmetric neighbours, showing in addition which observations belong
+to which subgraph. Obviously, no-neighbour observations are singleton
+graph nodes, but subgraphs are also troubling for spatial analysis,
+because there is no connection between the spatial processes in those
+subgraphs. The ripples in one pond cannot cross into a separate pond if
+they are not connected.
+
From spdep 1.3-1, steps began to raise awareness of the
+possibility that neighbour objects might be created that are
+disconnected in some way, mostly through warnings, and through the
+computation of subgraph measures by default. This vignette is intended
+to provide some background to these steps.
+
+
+
No-neighbour observations
+
+
From the start, nb objects have recorded no-neighbour
+observations as an integer vector of unit length and value
+0, where neighbours are recorded as ID values between
+1 and N, where N is the
+observation count. print and summary methods
+have always reported the presence of no-neighbour observations, and
+listed their IDs (or region.id values). If an
+nb object contains no-neighbour observations, the user has
+to decide whether to drop those observations, or if retained, what value
+to give its weights. The zero.policy argument uses zero as
+the value if TRUE, but if FALSE causes nb2listw to fail.
+The value of zero.policy in a call to functions like
+nb2listw, subset.listw or
+mat2listw creating listw objects representing
+sparse spatial weights matrices is added to the created object as an
+attribute, and used subsequently to pass through that choice to other
+functions. For example, moran.test takes the value of this
+attribute as default for its zero.policy argument:
If observation
+
+has no neighbours, its weights sum
+,
+as
+
+(see discussion in Bivand and Portnov
+(2004)). Its eigenvalue will also be zero, with consequences for
+analytical inference:
The adjust.n argument to measures of spatial
+autocorrelation is by default TRUE, and subtracts the count of singleton
+nodes from
+
+in an attempt to acknowledge the reduction in information available.
+
This discussion will address problems arising when analysing
+areal/lattice data, and neighbours are defined as polygon features with
+contiguous boundaries. One way in which no-neighbour observations may
+occur is when they are islands. This is clearly the case in Freni-Sterrantino, Ventrucci, and Rue (2018),
+where Capraia and Giglio Isles are singleton nodes. Here we take
+Westminster constituencies for Wales used in the July 2024 UK general
+election.
## Warning in poly2nb(w50m, row.names = as.character(w50m$Constituency)): some observations have no neighbours;
+## if this seems unexpected, try increasing the snap argument.
+
## Warning in poly2nb(w50m, row.names = as.character(w50m$Constituency)): neighbour object has 2 sub-graphs;
+## if this sub-graph count seems unexpected, try increasing the snap argument.
+
## Neighbour list object:
+## Number of regions: 32
+## Number of nonzero links: 136
+## Percentage nonzero weights: 13.28125
+## Average number of links: 4.25
+## 1 region with no links:
+## Ynys Môn
+## 2 disjoint connected subgraphs
+
The two subgraphs are the singleton Ynys Môn and all the other 31
+constituencies:
The left map shows that Ynys Môn can be shown selecting by name, as a
+black border, and by the zero cardinality of its neighbour set, using
+card, filling the polygon. The right map shows the location
+of the island, known in English as Anglesey, north-west of the Welsh
+mainland, and with no neighbour links:
From the maps, we can see that the island is close to two
+constituencies across the Afon Menai (Menai Strait in English), the
+three simplified polygons being less than 280m apart, measured between
+polygon boundaries:
## Neighbour list object:
+## Number of regions: 32
+## Number of nonzero links: 140
+## Percentage nonzero weights: 13.67188
+## Average number of links: 4.375
In this case, increasing snap from its default of 10mm
+(or close equivalents for geometries with known metrics; previously
+sqrt(.Machine$double.eps) in all cases) helps. The
+symmetric links added are to:
This is not always going to be the case, but here the strait is
+narrow. If islands are much further offshore, other steps may be
+required, because a large snap distance will draw in extra
+neighbours for already connected observations. It is also possible that
+increasing the snap distance may fail to link islands if
+they are not considered candidate neighbours, that is if their extents
+(bounding boxes), buffered out by the snap value, do not
+intersect.
+
We can also use the distances to pick out those neighbour candidates
+that meet our criterion of 280m, taking care not to lose the ordering
+needed to identify the correct observations:
The addlinks1 function can be used to add both the links
+from Ynys Môn to its neighbours, and by symmetry from them to Ynys Môn.
+This approach means that each island should be treated separately (or
+scripted in sequence), but does not risk adding spurious neighbours in
+denser parts of the study area.
+
+(nb_W_50m_add<-addlinks1(nb_W_50m, from =cands[1], to =cands[2:meet_criterion]))
+
## Neighbour list object:
+## Number of regions: 32
+## Number of nonzero links: 140
+## Percentage nonzero weights: 13.67188
+## Average number of links: 4.375
Since these constituency observations have areal support, it is not
+surprising that changing support to points and using
+-nearest
+neighbours does not work adequately, because the distance measurements
+are between the points representing the polygons rather than as above
+between the areal unit boundaries:
Here, Clwyd North, east of Bangor Aberconwy, is given as a neighbour
+of Ynys Môn but Dwyfor Meirionnydd, west of Bangor Aberconwy, is not. In
+addition, there are two subgraphs, which still remain up to
+.
+
+
+
Subgraphs
+
+
Subgraphs may be found when no-neighbour observations are present,
+but also when the graph is split between two blocks of observations with
+no path from any observation in a block to any in another block, across
+the low population density constituencies in mid-Wales:
## Warning in knn2nb(knearneigh(pts, k = 6), row.names =
+## as.character(w50m$Constituency), : neighbour object has 2 sub-graphs
+
## Neighbour list object:
+## Number of regions: 32
+## Number of nonzero links: 238
+## Percentage nonzero weights: 23.24219
+## Average number of links: 7.4375
+## 2 disjoint connected subgraphs
This occurs frequently with point support, but may also occur with
+areal support, as Freni-Sterrantino, Ventrucci,
+and Rue (2018) find for the eight municipalities on the island of
+Elba.
+
From spdep 1.3-6, if the igraph and
+spatialreg packages are available, n.comp.nb
+uses igraph::components to compute the graph components,
+also using depth-first search. The original implementation is as fast,
+but for directed (asymmetric) graphs converts first to symmetry, while
+igraph::components can handle directed graphs without such
+conversion (see https://github.com/r-spatial/spdep/issues/160 for
+details).
## Warning in knn2nb(knearneigh(pts, k = 6), row.names =
+## as.character(w50m$Constituency)): neighbour object has 2 sub-graphs
+
## Neighbour list object:
+## Number of regions: 32
+## Number of nonzero links: 192
+## Percentage nonzero weights: 18.75
+## Average number of links: 6
+## 2 disjoint connected subgraphs
+## Non-symmetric neighbours list
+
Another case demonstrates how cyclical subgraphs may appear; this is
+again taken from constituencies in the 2024 UK general election,
+subsetted to those in England south of London.
## Warning in poly2nb(sc50m, row.names = as.character(sc50m$Constituency)): neighbour object has 2 sub-graphs;
+## if this sub-graph count seems unexpected, try increasing the snap argument.
+
## Neighbour list object:
+## Number of regions: 119
+## Number of nonzero links: 530
+## Percentage nonzero weights: 3.742674
+## Average number of links: 4.453782
+## 2 disjoint connected subgraphs
+
The second subgraph only has two members, who are each others’ only
+neighbours, known as a cyclical component.
This has consequences for the eigenvalues of the spatial weights
+matrix, pointed out by Smirnov and Anselin
+(2009) and Bivand, Hauke, and Kossowski
+(2013). With row-standardised weights, the eigenvalues of this
+component are:
Using 5km as a cutoff seems prudent, but would not work as a
+snap value. Taking Isle of Wight East first, there are four
+constituencies with boundaries within 5km:
## [1] "Isle of Wight East" "Isle of Wight West" "Portsmouth South"
+## [4] "Gosport"
+
+(nb_sc_50m_iowe<-addlinks1(nb_sc_50m, from =cands[1], to =cands[3:meet_criterion]))
+
## Neighbour list object:
+## Number of regions: 119
+## Number of nonzero links: 534
+## Percentage nonzero weights: 3.77092
+## Average number of links: 4.487395
+
Although all constituencies are now linked, we should see whether
+using the 5km criterion brings in extra neighbours for Isle of Wight
+West:
It, does, but we need to beware of the sorting order of the zero
+self-distance and contiguous neighbour distance, where from
+is now in the second position:
## [1] "Isle of Wight East" "Isle of Wight West" "New Forest West"
+## [4] "New Forest East"
+
This then yields links to the north-west:
+
+(nb_sc_50m_iow<-addlinks1(nb_sc_50m_iowe, from =cands[2], to =cands[3:meet_criterion]))
+
## Neighbour list object:
+## Number of regions: 119
+## Number of nonzero links: 538
+## Percentage nonzero weights: 3.799167
+## Average number of links: 4.521008
It remains to add a suitable generalisation of addlinks1
+to handle a from vector argument and a to
+argument taking a list of vectors.
+
+
+
Unintentional disconnected graphs
+
+
Sometimes apparently sensible polygons turn out to be represented in
+such a way that disconnected graphs are generated when extracting
+contiguities. One such case was raised in https://github.com/r-spatial/spdep/issues/162, for
+subdivisions of Tokyo.
## Warning in poly2nb(tokyo, snap = sqrt(.Machine$double.eps)): some observations have no neighbours;
+## if this seems unexpected, try increasing the snap argument.
+
## Warning in poly2nb(tokyo, snap = sqrt(.Machine$double.eps)): neighbour object has 23 sub-graphs;
+## if this sub-graph count seems unexpected, try increasing the snap argument.
+
## Neighbour list object:
+## Number of regions: 262
+## Number of nonzero links: 946
+## Percentage nonzero weights: 1.378125
+## Average number of links: 3.610687
+## 10 regions with no links:
+## 101, 127, 134, 135, 152, 154, 167, 237, 242, 243
+## 23 disjoint connected subgraphs
+
The legacy default snap value when the coordinates are
+measured in metres was 15 nanometres, which effectively assumed that the
+coordinates making up polygon boundaries were identical:
## Neighbour list object:
+## Number of regions: 262
+## Number of nonzero links: 1390
+## Percentage nonzero weights: 2.02494
+## Average number of links: 5.305344
On that basis, the default was changed from spdep 1.3-6
+to 10mm for projected polygons, and the snap value used was returned as
+an attribute of the neighbour object:
## Neighbour list object:
+## Number of regions: 262
+## Number of nonzero links: 1390
+## Percentage nonzero weights: 2.02494
+## Average number of links: 5.305344
## Neighbour list object:
+## Number of regions: 262
+## Number of nonzero links: 1338
+## Percentage nonzero weights: 1.949187
+## Average number of links: 5.10687
+
The default snap value used in poly2nb when
+the polygons are expressed in decimal degrees is:
This was set based on the apparent “size” of 10mm in decimal
+degrees:
+
+(180*0.01)/(pi*6378137)
+
## [1] 8.983153e-08
+
+
+
References
+
+
+
+Bivand, Roger, Jan Hauke, and Tomasz Kossowski. 2013. “Computing
+the Jacobian in Gaussian Spatial
+Autoregressive Models: An Illustrated Comparison of Available
+Methods.”Geographical Analysis 45 (2): 150–79. https://doi.org/10.1111/gean.12008.
+
+
+Bivand, Roger, and B. A. Portnov. 2004. “Exploring Spatial Data
+Analysis Techniques Using R: The Case of Observations with
+No Neighbours.” In Advances in Spatial Econometrics:
+Methodology, Tools, Applications, edited by Luc Anselin, Raymond J.
+G. M. Florax, and S. J. Rey, 121–42. Berlin: Springer. https://doi.org/10.1007/978-3-662-05617-2_6.
+
+
+Freni-Sterrantino, Anna, Massimo Ventrucci, and Håvard Rue. 2018.
+“A Note on Intrinsic Conditional Autoregressive Models for
+Disconnected Graphs.”Spatial and Spatio-Temporal
+Epidemiology 26: 25–34. https://doi.org/10.1016/j.sste.2018.04.002.
+
+
+Smirnov, O., and L. Anselin. 2009. “An O(N) Parallel
+Method of Computing the Log-Jacobian of the
+Variable Transformation for Models with Spatial Interaction on a
+Lattice.”Computational Statistics & Data Analysis
+53 (8): 2980–88. https://doi.org/10.1016/j.csda.2008.10.010.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff --git a/docs/articles/subgraphs_files/figure-html/unnamed-chunk-10-1.png b/docs/articles/subgraphs_files/figure-html/unnamed-chunk-10-1.png
new file mode 100644
index 00000000..73eb45c3
Binary files /dev/null and b/docs/articles/subgraphs_files/figure-html/unnamed-chunk-10-1.png differ
diff --git a/docs/articles/subgraphs_files/figure-html/unnamed-chunk-18-1.png b/docs/articles/subgraphs_files/figure-html/unnamed-chunk-18-1.png
new file mode 100644
index 00000000..7b01a6ee
Binary files /dev/null and b/docs/articles/subgraphs_files/figure-html/unnamed-chunk-18-1.png differ
diff --git a/docs/articles/subgraphs_files/figure-html/unnamed-chunk-19-1.png b/docs/articles/subgraphs_files/figure-html/unnamed-chunk-19-1.png
new file mode 100644
index 00000000..07bac2a9
Binary files /dev/null and b/docs/articles/subgraphs_files/figure-html/unnamed-chunk-19-1.png differ
diff --git a/docs/articles/subgraphs_files/figure-html/unnamed-chunk-25-1.png b/docs/articles/subgraphs_files/figure-html/unnamed-chunk-25-1.png
new file mode 100644
index 00000000..7f173b24
Binary files /dev/null and b/docs/articles/subgraphs_files/figure-html/unnamed-chunk-25-1.png differ
diff --git a/docs/articles/subgraphs_files/figure-html/unnamed-chunk-37-1.png b/docs/articles/subgraphs_files/figure-html/unnamed-chunk-37-1.png
new file mode 100644
index 00000000..630e9f07
Binary files /dev/null and b/docs/articles/subgraphs_files/figure-html/unnamed-chunk-37-1.png differ
diff --git a/docs/articles/subgraphs_files/figure-html/unnamed-chunk-7-1.png b/docs/articles/subgraphs_files/figure-html/unnamed-chunk-7-1.png
new file mode 100644
index 00000000..536af3e3
Binary files /dev/null and b/docs/articles/subgraphs_files/figure-html/unnamed-chunk-7-1.png differ
diff --git a/docs/authors.html b/docs/authors.html
index 382ff10a..d188f3f1 100644
--- a/docs/authors.html
+++ b/docs/authors.html
@@ -43,6 +43,9 @@
coords<-st_coordinates(st_centroid(st_geometry(columbus)))plot(col.gal.nb, coords, col="grey")col2<-droplinks(col.gal.nb, 21)
+#>Warning: some observations have no neighbours#>Warning: neighbour object has 3 sub-graphsres<-n.comp.nb(col2)table(res$comp.id)
@@ -213,12 +217,12 @@
Examples
system.time(udir<-n.comp.nb(make.sym.nb(k6)))}#> user system elapsed
-#> 2.238 0.000 2.252
+#> 2.435 0.000 2.457 if(run){system.time(dir<-n.comp.nb(k6))}#> user system elapsed
-#> 0.719 0.002 0.724
+#> 0.806 0.002 0.812 if(run){udir$nc}
diff --git a/docs/reference/diffnb.html b/docs/reference/diffnb.html
index b890c31f..3eb94951 100644
--- a/docs/reference/diffnb.html
+++ b/docs/reference/diffnb.html
@@ -43,6 +43,9 @@
#>Warning: neighbour object has 11 sub-graphssystem.time(o<-nbdists(gck1b, xy1))#> user system elapsed
-#> 0.006 0.000 0.006
+#> 0.007 0.000 0.007 (all.linked<-max(unlist(o)))#> [1] 522.4464# use s2 brute-force dwithin_matrix approach for s2 <= 1.0.7system.time(gc.nb.dwithin<-dnearneigh(xy1, 0, all.linked, use_s2=TRUE, dwithin=TRUE))#>Warning: neighbour object has 3 sub-graphs#> user system elapsed
-#> 0.012 0.000 0.012
+#> 0.013 0.000 0.014 summary(gc.nb, xy1, scale=0.5)#> Neighbour list object:#> Number of regions: 48
@@ -276,13 +279,13 @@
Examples
}#>Warning: neighbour object has 3 sub-graphs#> user system elapsed
-#> 0.01 0.00 0.01
+#> 0.012 0.000 0.011 if(packageVersion("s2")>"1.0.7"){system.time(gc.nb.dwithin<-dnearneigh(xy1, 0, all.linked, use_s2=TRUE, dwithin=TRUE))}#>Warning: neighbour object has 3 sub-graphs#> user system elapsed
-#> 0.011 0.000 0.011
+#> 0.013 0.000 0.013 if(packageVersion("s2")>"1.0.7"){summary(gc.nb.dwithin, xy1, scale=0.5)}
@@ -325,7 +328,7 @@
Examples
system.time(gc.nb.legacy<-dnearneigh(xy1, 0, all.linked, use_s2=FALSE))#>Warning: neighbour object has 3 sub-graphs#> user system elapsed
-#> 0.006 0.000 0.006
+#> 0.007 0.000 0.007 summary(gc.nb, xy1, scale=0.5)#> Neighbour list object:#> Number of regions: 48
diff --git a/docs/reference/droplinks.html b/docs/reference/droplinks.html
index 2a7ce5f2..a346d1b8 100644
--- a/docs/reference/droplinks.html
+++ b/docs/reference/droplinks.html
@@ -1,5 +1,5 @@
-Drop links in a neighbours list — droplinks • spdepDrop and add links in a neighbours list — droplinks • spdep
@@ -43,6 +43,9 @@
Drops links to and from or just to a region from a neighbours list. The example corresponds to Fingleton's Table 1, p. 6, for lattices 5 to 19.
+
droplinks drops links to and from or just to a region from a neighbours list. The example corresponds to Fingleton's Table 1, (1999) p. 6, for lattices 5 to 19. addlinks1 adds links from a single region to specified regions.
-
droplinks(nb, drop, sym=TRUE)
+
droplinks(nb, drop, sym=TRUE)
+addlinks1(nb, from, to, sym=TRUE)
@@ -85,7 +89,13 @@
Arguments
either a logical vector the length of nb, or a character vector of named regions corresponding to nb's region.id attribute, or an integer vector of region numbers
sym
-
TRUE for removal of both "row" and "column" links, FALSE for only "row" links
+
TRUE for removal of both "row" and "column" links, FALSE for only "row" links; when adding links, inserts links to the from region from the to regions
+
+
from
+
single from region for adding links, either a character vector of length 1 of the named from region corresponding to nb's region.id attribute, or an integer vector of length 1 holding a region number
+
+
to
+
to regions, either a character vector of named from regions corresponding to nb's region.id attribute, or an integer vector of region numbers
@@ -128,6 +138,14 @@
Examples
mxs[i,j]<-max(res)}}
+#>Warning: some observations have no neighbours
+#>Warning: some observations have no neighbours
+#>Warning: some observations have no neighbours
+#>Warning: some observations have no neighbours
+#>Warning: some observations have no neighbours
+#>Warning: some observations have no neighbours
+#>Warning: some observations have no neighbours
+#>Warning: some observations have no neighboursprint(mns)#> 0.2 0.5 0.95 0.999 1#> 5 1.038271 1.312627 9.486051 30.81487 32.04915
diff --git a/docs/reference/edit.nb.html b/docs/reference/edit.nb.html
index 7f2a69f7..a174c1ac 100644
--- a/docs/reference/edit.nb.html
+++ b/docs/reference/edit.nb.html
@@ -48,6 +48,9 @@
#> Percentage nonzero weights: 7.080383 #> Average number of links: 3.469388 #> 4 regions with no links:
-#> 1005 1006 1008 1043
+#> 1005, 1006, 1008, 1043#> 8 disjoint connected subgraphs#> Link number distribution:#>
@@ -159,7 +162,7 @@
Examples
#> Percentage nonzero weights: 7.080383 #> Average number of links: 3.469388 #> 4 regions with no links:
-#> 1005 1006 1008 1043
+#> 1005, 1006, 1008, 1043#> 8 disjoint connected subgraphs#> Link number distribution:#>
@@ -177,10 +180,11 @@
#> Percentage nonzero weights: 2.360128 #> Average number of links: 3.469388 #> 12 regions with no links:
-#> 1 3 6 21 50 52 55 70 99 101 104 119
+#> 1, 3, 6, 21, 50, 52, 55, 70, 99, 101, 104, 119#> 24 disjoint connected subgraphs#> Link number distribution:#>
@@ -229,7 +233,7 @@
Examples
#> Percentage nonzero weights: 7.080383 #> Average number of links: 3.469388 #> 4 regions with no links:
-#> 1005 1006 1008 1043
+#> 1005, 1006, 1008, 1043#> 8 disjoint connected subgraphs#> Link number distribution:#>
@@ -249,10 +253,11 @@
#> Max. :0.5000 tf<-paste0(tempfile(), ".gpkg")st_write(res, dsn=tf, driver="GPKG")
-#> Writing layer `file71d532de9e00d' to data source
-#> `/tmp/Rtmp6barmj/file71d532de9e00d.gpkg' using driver `GPKG'
+#> Writing layer `filec1856d3811f' to data source
+#> `/tmp/Rtmp3CAEPn/filec1856d3811f.gpkg' using driver `GPKG'#> Writing 230 features with 5 fields and geometry type Line String.inMap<-st_read(tf)
-#> Reading layer `file71d532de9e00d' from data source
-#> `/tmp/Rtmp6barmj/file71d532de9e00d.gpkg' using driver `GPKG'
+#> Reading layer `filec1856d3811f' from data source
+#> `/tmp/Rtmp3CAEPn/filec1856d3811f.gpkg' using driver `GPKG'#> Simple feature collection with 230 features and 5 fields#> Geometry type: LINESTRING#> Dimension: XY
@@ -192,8 +195,9 @@
all(isTRUE(all.equal(cal1a_1n$neighbours, cal1_1n_rt$neighbours)))all(isTRUE(all.equal(cal1a_1n$weights, cal1_1n_rt$weights)))}
+}#> Reading layer `nc.gpkg' from data source `/home/rsb/lib/r_libs/sf/gpkg/nc.gpkg' using driver `GPKG'#> Simple feature collection with 100 features and 14 fields#> Geometry type: MULTIPOLYGON
@@ -263,7 +273,7 @@
Examples
#>#>Warning: region.id not given, c(MYID, NID) range is 1:100#> Reading layer `california' from data source
-#> `/tmp/Rtmput1Tax/temp_libpath71c612b87ba64/spdep/etc/shapes/california.gpkg'
+#> `/tmp/RtmpYdUOF1/temp_libpathbff31ae05681/spdep/etc/shapes/california.gpkg.zip' #> using driver `GPKG'#> Simple feature collection with 58 features and 2 fields#> Geometry type: MULTIPOLYGON
@@ -279,6 +289,7 @@
Examples
#> no-neighbour observations found, set zero.policy to TRUE#>#>Warning: region.id not given, c(MYID, NID) range is 1:58
+#>Warning: some observations have no neighbours#>Warning: neighbour object has 4 sub-graphs#>Warning: neighbour object has 4 sub-graphs#>Warning: region.id not given, c(MYID, NID) range is 159:214
diff --git a/docs/reference/rotation.html b/docs/reference/rotation.html
index 011f8013..9a4b652c 100644
--- a/docs/reference/rotation.html
+++ b/docs/reference/rotation.html
@@ -43,6 +43,9 @@
#> [[1]]#> [1] FALSE#>
-#> [[1]]
-#> [1] TRUE
-#>
-#> [[1]]
-#> [1] FALSE
-#>
+#>Error in checkForRemoteErrors(lapply(cl, recvResult)): one node produced an error: there is no package called 'spdep'
par(mar=c(0,0,0,0))plot(st_geometry(bh), border=gray(.5))pts<-st_coordinates(st_centroid(bh))
-#>Warning: st_centroid assumes attributes are constant over geometriesplot(mst.bh, pts, col=2, cex.lab=.6, cex.circles=0.035, fg="blue", add=TRUE)
-### three groups with no restrictionres1<-skater(mst.bh[,1:2], dpad, 2)
-#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zero### groups sizetable(res1$groups)
-#>Error in eval(expr, envir, enclos): object 'res1' not found### the skater plotopar<-par(mar=c(0,0,0,0))plot(res1, pts, cex.circles=0.035, cex.lab=.7)
-#>Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'plot': object 'res1' not found### the skater plot, using other colorsplot(res1, pts, cex.circles=0.035, cex.lab=.7, groups.colors=heat.colors(length(res1$ed)))
-#>Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'plot': object 'res1' not found### the Spatial Polygons plotplot(st_geometry(bh), col=heat.colors(length(res1$edg))[res1$groups])
-#>Error in eval(expr, envir, enclos): object 'res1' not found
-par(opar)### EXPERT OPTIONS### more one partitionres1b<-skater(res1, dpad, 1)
-#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zero### length groups frequencytable(res1$groups)
-#>Error in eval(expr, envir, enclos): object 'res1' not foundtable(res1b$groups)
-#>Error in eval(expr, envir, enclos): object 'res1b' not found### thee groups with minimum population res2<-skater(mst.bh[,1:2], dpad, 2, 200000, bh$Pop)
-#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zerotable(res2$groups)
-#>Error in eval(expr, envir, enclos): object 'res2' not found### thee groups with minimun number of areasres3<-skater(mst.bh[,1:2], dpad, 2, 3, rep(1,nrow(bh)))
-#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zerotable(res3$groups)
-#>Error in eval(expr, envir, enclos): object 'res3' not found### thee groups with minimun and maximun number of areasres4<-skater(mst.bh[,1:2], dpad, 2, c(20,50), rep(1,nrow(bh)))
-#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zerotable(res4$groups)
-#>Error in eval(expr, envir, enclos): object 'res4' not found### if I want to get groups with 20 to 40 elementsres5<-skater(mst.bh[,1:2], dpad, 2,c(20,40), rep(1,nrow(bh)))## DON'T MAKE DIVISIONS
-#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zerotable(res5$groups)
-#>Error in eval(expr, envir, enclos): object 'res5' not found### In this MST don't have groups with this restrictions### In this case, first I do one division### with the minimun criteriares5a<-skater(mst.bh[,1:2], dpad, 1, 20, rep(1,nrow(bh)))
-#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zerotable(res5a$groups)
-#>Error in eval(expr, envir, enclos): object 'res5a' not found### and do more one division with the full criteriares5b<-skater(res5a, dpad, 1, c(20, 40), rep(1,nrow(bh)))
-#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zerotable(res5b$groups)
-#>Error in eval(expr, envir, enclos): object 'res5b' not found### and do more one division with the full criteriares5c<-skater(res5b, dpad, 1, c(20, 40), rep(1,nrow(bh)))
-#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zerotable(res5c$groups)
-#>Error in eval(expr, envir, enclos): object 'res5c' not found### It don't have another divison with this criteriares5d<-skater(res5c, dpad, 1, c(20, 40), rep(1,nrow(bh)))
-#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zerotable(res5d$groups)
-#>Error in eval(expr, envir, enclos): object 'res5d' not found# \dontrun{data(boston, package="spData")
@@ -297,16 +274,12 @@
Examples
dpad<-data.frame(scale(boston.c[,c(7:10)]))### calculating costssystem.time(lcosts<-nbcosts(bh.nb, dpad))
-#> user system elapsed
-#> 0.055 0.000 0.055 ### making listwnb.w<-nb2listw(bh.nb, lcosts, style="B")### find a minimum spanning treemst.bh<-mstree(nb.w,5)### three groups with no restrictionsystem.time(res1<-skater(mst.bh[,1:2], dpad, 2))
-#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zero
-#> Timing stopped at: 0 0 0library(parallel)nc<-max(2L, detectCores(logical=FALSE), na.rm =TRUE)-1L# set nc to 1L here
@@ -320,26 +293,24 @@
Examples
}### calculating costssystem.time(plcosts<-nbcosts(bh.nb, dpad))
-#> user system elapsed
-#> 0.043 0.000 0.044 all.equal(lcosts, plcosts, check.attributes=FALSE)
-#> [1] TRUE### making listwpnb.w<-nb2listw(bh.nb, plcosts, style="B")### find a minimum spanning treepmst.bh<-mstree(pnb.w,5)### three groups with no restrictionsystem.time(pres1<-skater(pmst.bh[,1:2], dpad, 2))
-#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zero
-#> Timing stopped at: 0 0 0if(!get.mcOption()){set.ClusterOption(NULL)stopCluster(cl)}all.equal(res1, pres1, check.attributes=FALSE)
-#>Error in h(simpleError(msg, call)): error in evaluating the argument 'target' in selecting a method for function 'all.equal': object 'res1' not foundinvisible(set.coresOption(coresOpt))# }
+}
+#>Warning: st_centroid assumes attributes are constant over geometries
+#>Error in if (w$num_obs < 1) { stop("The weights is not valid.")}: argument is of length zero
+
+We can show the block structure by displaying the binary spatial weights
+matrix:
