[R-sig-Geo] Adding a Few Neighbour Relationships to a nb List

[Barry Rowlingson]

I recently answered a similar question on Stack Overflow where someone
needed to add detached polygons to their connected network by connecting
them to their nearest neighbour:

https://stackoverflow.com/questions/57269254/how-to-impute-missing-neighbours-of-a-spatial-weight-matrix-queen-contiguity/57378930?noredirect=1#comment101246065_57378930

in short, you can treat a `nb` object like a list of vectors: nb[[i]] is a
vector of indexes of objects connected to object `i`

BUT you have to make sure you store integers:

Here's a `nb` object from that question which in summary has this
manyneighbours for each region:

> card(nb)
[1] 2 3 4 3 2 0 0

lets set the 6th feature to be a neighbour of the first:

> nb[[6]] = 1

then uh-oh...

> card(nb)
Error in card(nb) :
  INTEGER() can only be applied to a 'integer', not a 'double'

same again only `as.integer`:

> nb[[6]] = as.integer(1)

and its happy:

> card(nb)
[1] 2 3 4 3 2 1 0

if you want to set the nighbours of 6 to several features:

> nb[[6]] = as.integer(c(1,2,3))
> card(nb)
[1] 2 3 4 3 2 3 0

Barry



On Wed, Aug 7, 2019 at 10:25 PM Stuart Reece <stuart.reece using bigpond.com>
wrote:

> Thanks Vijay.
>
> Big help.
>
> I will go through the recommended chapter in detail.
>
> Normally for a list I can just use single square brackets like this [] to
> access the elements and change them by assignment.
>
> But I cannot quite figure out how to do this with the nb lists.
>
>
>
> Doing this with [[]] works really well to create nicely corrected graphs.
>
>
>
> But fails due it “out of bounds index errors” with regression equations????
>
>
>
> I find this ever so confusing….???
>
>
>
> Thankyou so much again,
>
>
>
> Stuart Reece.
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> From: Vijay Lulla <vijaylulla using gmail.com>
> Sent: Thursday, 8 August, 2019 1:53 AM
> To: Stuart Reece <stuart.reece using bigpond.com>
> Cc: Roger Bivand <Roger.Bivand using nhh.no>; R-sig-geo Mailing List <
> r-sig-geo using r-project.org>
> Subject: Re: [R-sig-Geo] Adding a Few Neighbour Relationships to a nb List
>
>
>
> Maybe
> https://cran.r-project.org/web/packages/spdep/vignettes/nb_igraph.html
> can help with all your questions.
> https://cran.r-project.org/web/packages/spdep/vignettes/nb_sf.html
> contains a little more detail about nb structure.  Finally, I encourage you
> to use `str` to study the structure of R objects.
>
> HTH,
>
> Vijay.
>
>
>
> On Wed, Aug 7, 2019 at 10:44 AM Stuart Reece <stuart.reece using bigpond.com
> <mailto:stuart.reece using bigpond.com> > wrote:
>
> Dear R-Sig-Geo list,
>
>
>
> I was wondering if it might be possible please to request assistance with
> adding some nb relationships to a .nb.gal list composed either by GeoDa or
> poly2nb in R????
>
>
>
> The shapefile at this URL
> <
> https://www.samhsa.gov/data/report/2014-2016-nsduh-substate-region-shapefil
> e>  divides USA into 395 substate regions.  For health and demographical
> reasons it is important to include both Hawaii and Alaska in the
> spatiotemporal analysis so I want to introduce these into the Southeast
> coast of California and the Pacific northwest respectively.
>
>
>
> This is just as Giovanni Millo added in spatial relationships for Sicily
> across the Strait of Messina for splm on page 7 of the splm pdf.
>
>
>
> I found edit.nb in spdep and operated it just as described in the
> instructions and here
> <https://github.com/r-spatial/spdep/blob/master/man/edit.nb.Rd> .  It
> crashed RStudio many times but ran well in R3.6.1.  However even though I
> assigned it to a new object it did not save well.  Although when I plotted
> the dxxx file as the difference between the old and modified files it
> plotted the changes beautifully in red and black respectively when plotted
> by themselves it introduced many long distance extraneous relationships.
> To
> get the edited nb list file out of R 3.6.1 and into RStudio I saved it as
> an
> RDS file.  However when opened in RStudio it was grossly erroneous and
> included extraneous links from Hawaii to Boston and New York.  When I
> opened
> the file in RStudio it again introduced these extraneous links.
>
>
>
> Saving it as a further new object in R 3.6.1 did not remedy these
> difficulties.
>
>
>
> The other problem I have is that the spdep poly2nb function excludes
> Richmond, an island off the southern tip of Long Island near New York as it
> is an island.  Also one of the areas - Region 10 in Washington DC - is also
> excluded for reasons of which I am unsure.
>
>
>
> I found some code here to just patch single areas
> <https://stat.ethz.ch/pipermail/r-sig-geo/2006-June/001073.html>  like
> this
> but when I run it, it throws an integer error
>
> "  INTEGER() can only be applied to a 'integer', not a 'double'
>
>
>
> No combination of bracketing around subscripts helps or works at all.  The
> link mentioned has these statements in it
>
>
>
> nb[[ij[1]]] <- sort(unique(c(nb[[ij[1]]], ij[2])))
> nb[[ij[2]]] <- sort(unique(c(nb[[ij[2]]], ij[1])))
>
>
>
> which makes me think that I should insert a vector "  c(i,j) "  where
> indicated.  Even using "c(as.integer(i),as.integer(j)) "
>
> or " as.integer(c(I,j)) "   doesn't work and still gives rise to the same
> error.
>
>
>
> I am sure I am not the only one to have encountered such difficulties but I
> have really tried everything I can think of.
>
>
>
> The other thing I would really like is some clear instructions as to the
> true underlying structure of the nb list.  If I could clearly understand
> this then I could just go into the affected lines of the list of lists and
> edit them directly.
>
>
>
> However I am quite unable to find any clear description of its structure on
> line.
>
>
>
> Similarly I cannot find the source code for drop.links online to try to
> translate this code into add.links directly, as was also suggested.
>
>
>
> But such a function would I think be enormously helpful and of invaluable
> assistance for final editing.
>
>
>
> Thankyou ever so much in advance for your kind and gracious assistance.
>
>
>
> Yours sincerely,
>
>
>
> Stuart Reece.
>
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo using r-project.org <mailto:R-sig-Geo using r-project.org>
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>
>
>
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo using r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>

	[[alternative HTML version deleted]]