On Thu, 21 Aug 2008, Massimo Ventrucci wrote:
Roger,
sorry if I reply so late to your, instead, really quick suggestion at my
problem, actually I revised the question and I am not yet able to do
that plot. I try to explain clearly my problem.
I need to plot a map, for instance, in the same way I can plot it with
winbugs from the mapping tool menu. So I need to plot a value for each
region of my map and to colour regions characterized by higher values. I
have the map in a Spatial Polygons object, I do not have it in shape
file unfortunately. So I am working with package "sp". Moreover I can
not use the spplot because I do not have a SpatialPolygonsDataFrame
object, just a SpatialPolygons. In other words I do not have a value for
each x-y coords, but I would need a value for each region, hence for
each polygon belonging to the list of polygons.
Say SP is your SpatialPolygons object, and var is a numeric vector. Is
length(var) == length(slot(SP, "polygons"))
Are the values in var ordered in the same order as the polygons in SP? If
so, you may define the class interval break points and a colour palette as
say:
brks <- quantiles(var, seq(0, 1, 1/7))
cols <- grey.colors(length(brks)-1, 0.4, 0.9)
plot(SP, col=cols[findInterval(var, brks, all.inside=TRUE))
If the lengths agree, but the order doesn't, look at and identify the ID
values in the SpatialPolygons object:
IDs <- sapply(slot(SP, "polygons"), function(x) slot(x, "ID"))
plot(SP)
text(coordinates(SP), labels=IDs, cex=0.7)
Next add the matching IDs values to the values in var as a vector of
names, and:
names(var) <- IDs_in_correct_order
SPDF <- SpataialPolygonsDataFrame(SP, as.data.frame(var)))
to make a SpatialPolygonsDataFrame. Then you can either:
spplot(SPDF, "var")
or
plot(SPDF, col=cols[findInterval(SPDF$var, brks, all.inside=TRUE))
(untried but should work).
Hope this helps,
Roger
Please, any suggestion about which is the easiest way to do it, if any?
Thank you very much in advance
- massimo
Massimo Ventrucci
Ph.D Student in Statistics
Dipartimento di Scienze Statistiche
Università degli studi di Bologna
Via Belle Arti 41
40126 Bologna-Italy
_____________________________________
Phone +39(051)264182 (office)
Fax +39(051)232153
E-mail [EMAIL PROTECTED]
On Mon, 28 Jul 2008, Massimo Ventrucci wrote:
Dear all,
sorry my question seems to me quite basic and perhaps is already posted
but at a first sight I did not find it. Basically I have a map of
contiguous areas in the form of a SpatialPolygons object. In all areas I
need to display the value of a binary value (e.g. to color the areas
corresponding to value 1). I have to say in a guide to the package
"spdep" I found the code to do a map (it was a map of probabilities
buillt by using plot coomand after having used probmap command), but in
that case the plot is done starting from an object of class polylist if
I say well. Instead I need to do a plot starting from a SpatialPolygons
object because I do not have the map in other formats.
I'm planning to update the now very obsolete vignette in spdep soon. The
easiest way to plot a variable from a SpatialPolygonsDataFrame object is
the spplot method in the sp package (which spdep loads). Just try:
spplot(my_SpolDF, "myvar")
for a first cut.
Base graphics methods, such as plot(), are also available. The best online
guide may be one of the units at:
http://www.bias-project.org.uk/ASDARcourse/
either "Vizualising Spatial Data" for the description, or your choice of
the worked examples. There are also code examples and figures on the ASDAR
book website: www.asdar-book.org; though this is still under construction,
most things are there.
Roger
Moreover, in future I will also need to plot values graduated by colors.
Please, can you help me?
thanks in advance
massimo
Massimo Ventrucci
Ph.D Student in Statistics
Dipartimento di Scienze Statistiche
Università degli studi di Bologna
Via Belle Arti 41
40126 Bologna-Italy
_____________________________________
Phone +39(051)264182 (office)
Fax +39(051)232153
E-mail massimo.ventrucci at unibo.it
<https://stat.ethz.ch/mailman/listinfo/r-sig-geo>
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo at stat.math.ethz.ch <https://stat.ethz.ch/mailman/listinfo/r-sig-geo>
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
--
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: [EMAIL PROTECTED]
_______________________________________________
R-sig-Geo mailing list
[email protected]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
