R help - Feb 2004 - Data for use in maps()

Dear all,

I am interested in plotting maps visualizing spatial statistics in an
aggregated fashion, according to administrative boundaries. 
More specifically, I have fitted a cross-section model on data regarding
Italian "counties" (province, for Italian readers) and I would like to
visualize residual behavior on a map, in order to have a first
assessment of their spatial autocorrelation. I would also make some EDA
on the spatial patterns (if any) of the regressors.

I have found the maps package (and related) and would be able to do what
I want, e.g., for the USA, essentially by
>map("state",fill=T,col=color)
where color is dependent on the statistic of interest, but I still lack
a data file for counties' boundaries in Italy. Does anybody know where
to find one? Is there any convenient tool for converting from other
formats? I would like to do everything in R if possible.

Thanks in advance

Giovanni Millo
R&D Dept.
Assicurazioni Generali SpA
Trieste, Italy
Ai sensi del D.Lgs.196/2003 si precisa che le informazioni contenute in questo
messaggio sono riservate ed a uso esclusivo del destinatario. Qualora il
messaggio in parola Le fosse pervenuto per errore, La invitiamo ad eliminarlo
senza copiarlo e a non inoltrarlo a terzi, dandocene gentilmente comunicazione.
Grazie.<BR><BR>Pursuant to Legislative Decree No. 196/2003, you are
hereby informed that this message contains confidential information intended
only for the use of the addressee. If you are not the addressee, and have
received this message by mistake, please delete it and immediately notify us.
You may not copy or disseminate this message to anyone. Thank you.
	[[alternative HTML version deleted]]

On Mon, 16 Feb 2004, Millo Giovanni wrote:
> Dear all,
> 
> I am interested in plotting maps visualizing spatial statistics in an
> aggregated fashion, according to administrative boundaries. 
> More specifically, I have fitted a cross-section model on data regarding
> Italian "counties" (province, for Italian readers) and I would
like to
> visualize residual behavior on a map, in order to have a first
> assessment of their spatial autocorrelation. I would also make some EDA
> on the spatial patterns (if any) of the regressors.
> 
> I have found the maps package (and related) and would be able to do what
> I want, e.g., for the USA, essentially by
> >map("state",fill=T,col=color)
> where color is dependent on the statistic of interest, but I still lack
> a data file for counties' boundaries in Italy. Does anybody know where
> to find one? Is there any convenient tool for converting from other
> formats? I would like to do everything in R if possible.
> 
Unfortunately, while the US government makes available a large amount of 
free map data - such as administrative boundaries, this is certainly not 
the case in Europe. Of course, administrative boundaries also change, so 
the map data source you use should match the data you are trying to map. 
At least two proprietary GIS programs distribute maps of Italian provinces 
for use in their programs (MapInfo and ArcGIS), but I believe you need a 
license to access the data - it is (unlike US data) not downloadable.

It is possible that you can use the data referenced in the RArcInfo 
contributed package, which is downloaded from:

http://www.grid.unep.ch/data/grid/gnv158.php

but these are NUTS 2 regions dating at latest from the early 1990's, and 
do not seem to be the same as the CIA Italian provinces map. 

So there are three questions:

1) finding a free (or non-free) source of boundary data appropriate for
the provinces that suit your attribute data;

2) getting that into R (and possibly projecting it);

3) plotting the map.

On 2 and 3, you may find http://sal.agecon.uiuc.edu/csiss/Rgeo/index.html 
useful, especially the sections on maps. It may be that your organisation 
has purchased suitable non-free boundary data - in which case a 
"shapefile" is probably the easiest format to ask for.
> Thanks in advance
> 
> Giovanni Millo
> R&D Dept.
> Assicurazioni Generali SpA
> Trieste, Italy
-- 
Roger Bivand
Econonic Geography Section, Department of Economics, Norwegian School of 
Economics and Business Administration, Breiviksveien 40, N-5045 Bergen, 
Norway, voice: +47-55959355, fax: +47-55959393; Roger.Bivand at nhh.no

> I am interested in plotting maps visualizing spatial statistics in an
> aggregated fashion, according to administrative boundaries.
> More specifically, I have fitted a cross-section model on data
regarding
> Italian "counties" (province, for Italian readers) and I would
like to
> visualize residual behavior on a map, in order to have a first
> assessment of their spatial autocorrelation. I would also make some
EDA
> on the spatial patterns (if any) of the regressors.
>
> I have found the maps package (and related) and would be able to do
what
> I want, e.g., for the USA, essentially by
> >map("state",fill=T,col=color)
> where color is dependent on the statistic of interest, but I still
lack
> a data file for counties' boundaries in Italy. Does anybody know where
> to find one? Is there any convenient tool for converting from other
> formats? I would like to do everything in R if possible.
>
> Thanks in advance
>
> Giovanni Millo
> R&D Dept.
Dear Millo Giovanni:

Attached is a zip file containing three files.  The file
'italy.prov.pol'
is a R/Splus format polygon file of the provinces.  In this file, there
are NA records in the $y column where the polygons end.  The
corresponding
$x column are polygon identifiers.

The file 'italy.prov.pat' has three columns: the first are the polygon
identifiers used in 'italy.prov.pol', the second are the NUTS3
identifiers,
and the third are the names.  The file 'italy.prov.crf' has commas
separating the fields.

The file italy.prov.e00 is an arc/info export file of a polygon coverage
of the province boundaries.  This was extracted from the European NUTS3
file of third level political boundaries available at:
http://www.grid.unep.ch/data/grid/gnv159.php.  The R/S format files
above
were made from this arc/info file.

All location coordinates in the above files are geographic
(longitude, latitude).

In Becker and Wilks document "Constructing a Geographical Database"
that
accompanies the 'maps' package there is a description of how to make a
database that can be used by the maps package.  I am not aware that
anyone
has made this capability available in R, and I have not done this
either.

You can read the .pol file like this:

prov <- read.table ("italy.prov.pol", header=TRUE)

You can then use this function:

plot.map <- function (x, y) {
  rx <- range (x[!is.na(y)], na.rm=TRUE)
  ry <- range (y[!is.na(x)], na.rm=TRUE)
  plot.new ()
  plot.window (rx, ry, asp=1)
  }

and say

plot.map (prov$x, prov$y)
polygon (prov)

You can read the .pat file like this:

pat <- read.table ("italy.prov.pat", sep=",",
header=TRUE, quote="")

With this file you can connect the polygon boundaries in the .pol file
with names in .pat and with your statistics.  There is a crude mapping
facility available in contributed package "maptree".

Best wishes,

Denis White
   US EPA, 200 SW 35th St, Corvallis, Oregon, 97333 USA
   voice: 541.754.4476, email: white.denis at epa.gov
   web: www.epa.gov/wed/pages/staff/white/


< attachment deleted>