https://stat.ethz.ch/mailman/listinfo/r-sig-geo
or, via email, send a message with subject or body 'help' to
r-sig-geo-request at stat.math.ethz.ch
You can reach the person managing the list at
r-sig-geo-owner at stat.math.ethz.ch
When replying, please edit your Subject line so it is more specific
than "Re: Contents of R-sig-Geo digest..."
Today's Topics:
1. Re: Spatially Constrained Clustering (Elias T. Krainski)
2. regression kriging in gstat with skewed distributions (G. Allegri)
3. I Would Dream (vacates at patiencegroup.com)
4. R from cgi and Xvfb (G. Allegri)
5. Re: regression kriging in gstat with skewed distributions
(Tomislav Hengl)
----------------------------------------------------------------------
Message: 1
Date: Tue, 15 Jan 2008 10:51:48 -0300 (ART)
From: "Elias T. Krainski" <eliaskrainski at yahoo.com.br>
Subject: Re: [R-sig-Geo] Spatially Constrained Clustering
To: r-sig-geo at stat.math.ethz.ch
Message-ID: <569895.79461.qm at web50604.mail.re2.yahoo.com>
Content-Type: text/plain; charset=iso-8859-1
Hello Carson,
See the SKATER software at
http://www.est.ufmg.br/leste/skater.htm
The SKATER is a Spatial 'K'luster Analisys by Tree
Edge Removal. In future, this method also be available
in R.
Best,
Elias.
--- Carson Farmer <cfarmer at uvic.ca> escreveu:
Hello List,
I am trying to find an R package that will
accommodate spatially
constrained clustering. While I have been unable to
find a package that
is explicitly designed to do spatially constrained
clustering, I was
wondering if anyone had found a package that would
do constrained
clustering of any kind, and adapted this to spatial
constraints?
I have searched the R site extensively, and googled
all night long, but
to no avail! I HAVE found this post:
http://finzi.psych.upenn.edu/R/Rhelp02a/archive/56819.html
but the replies did not help much. They lead to
several packages which
perform spatial clustering (such that significant
clusters of say a
disease are located within a study region), however,
what I would like
to do is partition a spatial (grid) dataset based on
multiple variables,
taking into account their spatial locations (i.e.
clustering is based on
the variables, but constrained so that clusters are
spatially
contiguous). I'm thinking mclust is probably the
best way to go, but
I'm not sure where to start.
Any suggestions would be greatly appreciated.
Thanks,
Carson
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo at stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Elias T. Krainski
------------------------------
Message: 2
Date: Tue, 15 Jan 2008 15:27:58 +0100
From: "G. Allegri" <giohappy at gmail.com>
Subject: [R-sig-Geo] regression kriging in gstat with skewed
distributions
To: r-sig-geo at stat.math.ethz.ch
Message-ID:
<e12429640801150627q7266a599g77a3df9744edb153 at mail.gmail.com>
Content-Type: text/plain; charset=WINDOWS-1252
I'm trying to realize e regression kriging with gstat package on my
soil samples data. The response variable (ECe measuere) and covariates
appear positvely skewed.
As Tomislav Hengl suggests in its "framework for RK" [1], a logistic
transformation is proposed as a generic way to reduce the skeweness by
using the physical limits of the data.
Is it really a transformation that can be applied in the generic case
of skewed datas? I mean,in my case I have non-normal residuals (from
original data regression), and I'm trying to transform the residuals
(and not the original values) to do SK on them . Is this approach
correct?
A related question is how to do normal score transformations (for my
residuals) in R and gstat. I know gstat doesn't manage transformations
and back-transformations, so it should be done previously in R... but
I can't find any package that permit it in a straisghtforward way.
I've found something with qqnorm(ppoints(data)) and the approx()
function. Is that all?
Giovanni
[1] "A generic framework for spatial prediction of soil variables
based on regressionkriging" Geoderma 122 (1?2), 75?93.
------------------------------
Message: 3
Date: Tue, 15 Jan 2008 19:25:16 +0100
From: <vacates at patiencegroup.com>
Subject: [R-sig-Geo] I Would Dream
To: r-sig-geo at stat.math.ethz.ch
Message-ID: <478CFA8C.7030508 at patiencegroup.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Kisses Through E-mail http://86.123.21.76/
------------------------------
Message: 4
Date: Wed, 16 Jan 2008 11:04:53 +0100
From: "G. Allegri" <giohappy at gmail.com>
Subject: [R-sig-Geo] R from cgi and Xvfb
To: r-sig-geo at stat.math.ethz.ch
Message-ID:
<e12429640801160204h3532ddb9t624b4f6ef658c975 at mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Hi everyone.
I'm sorry for the question maybe OT.
I'm trying to use R and Python to run some scripts via web interface.
I've successfully setup mod_python for Apache and the rpy module.
R needs X11 to use png() and jpeg() devices, so I have installed Xvfb
(X virtual framebuffer). I works correctly: if I set the DISPLAY
variable to point to this X server, rpy can create png files correctly
from command-line, but it doesn't work when the python script is run
from web browser.
I restarted Apache after setting the DISPLAY variable, but the
Traceback gives me always the same error, about being not able to open
the X11 device?
Does anyone have made it work right?
How can tell Apache to run R script and forwarding X requests to my
Xvfb.
Thanks,
Giovanni
------------------------------
Message: 5
Date: Wed, 16 Jan 2008 11:08:28 +0100
From: "Tomislav Hengl" <hengl at science.uva.nl>
Subject: Re: [R-sig-Geo] regression kriging in gstat with skewed
distributions
To: "'G. Allegri'" <giohappy at gmail.com>
Cc: r-sig-geo at stat.math.ethz.ch
Message-ID: <001401c85827$bdac65a0$3a871291 at pcibed193>
Content-Type: text/plain; charset="windows-1250"
Dear Giovanni,
Logit transformation can be automatically applied to any variables which
has a lower and upper
physical limits (e.g. 0-100%). In R, you can transform a variable to
logits by e.g.:
points = read.dbf("points.dbf")
points$SANDt = log((points$SAND/100)/(1-(points$SAND/100)))
After you interpolate your variable, you can back-transform the values
by using:
SAND.rk = krige(fsand$call$formula, points[sel,], SPC, sand.rvgm)
SAND.rk$pred=exp(SAND.rk$var1.pred)/(1+exp(SAND.rk$var1.pred))*100
The prediction variance can not be back-transformed, but you can use the
normalized prediction
variance by dividing it with the sampled variance. See also section
4.2.1 of my lecture notes
(http://geostat.pedometrics.org/).
There are many transformations that can be applied to force a normality
of your target variable (see
e.g. http://en.wikipedia.org/wiki/Data_transformation_(statistics) ).
The most generic
transformation is to work with the probability density function values
(see e.g.
http://dx.doi.org/10.1016/j.jneumeth.2006.11.004 ), this way you do not
have to think about how the
histogram looks at all. But then the interpretation of the regression
plots becomes rather
difficult.
In any case, you should apply the transformation already to the target
variable because also a
requirement for linear regression is that the residuals are normally
distributed around the
regression line.
see also:
FITTING DISTRIBUTIONS WITH R (by Vito Ricci)
http://cran.r-project.org/doc/contrib/Ricci-distributions-en.pdf
Tom Hengl
http://spatial-analyst.net
-----Original Message-----
From: r-sig-geo-bounces at stat.math.ethz.ch
[mailto:r-sig-geo-bounces at stat.math.ethz.ch] On Behalf Of
G. Allegri
Sent: dinsdag 15 januari 2008 15:28
To: r-sig-geo at stat.math.ethz.ch
Subject: [R-sig-Geo] regression kriging in gstat with skewed
distributions
I'm trying to realize e regression kriging with gstat package on my
soil samples data. The response variable (ECe measuere) and covariates
appear positvely skewed.
As Tomislav Hengl suggests in its "framework for RK" [1], a logistic
transformation is proposed as a generic way to reduce the skeweness by
using the physical limits of the data.
Is it really a transformation that can be applied in the generic case
of skewed datas? I mean,in my case I have non-normal residuals (from
original data regression), and I'm trying to transform the residuals
(and not the original values) to do SK on them . Is this approach
correct?
A related question is how to do normal score transformations (for my
residuals) in R and gstat. I know gstat doesn't manage transformations
and back-transformations, so it should be done previously in R... but
I can't find any package that permit it in a straisghtforward way.
I've found something with qqnorm(ppoints(data)) and the approx()
function. Is that all?
Giovanni
[1] "A generic framework for spatial prediction of soil variables
based on regressionkriging" Geoderma 122 (1?2), 75?93.
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo at stat.math.ethz.ch
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
End of R-sig-Geo Digest, Vol 53, Issue 14
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo at stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-sig-geo