Marc Schwartz [EMAIL PROTECTED] writes:
In checking some of the other popular slide classes (ie. Seminar,
Prosper and Beamer), I don't think that they support the figure
environment either.
The figure environment is better for longer documents, where it can
become difficult to keep track
Hallo,
I want to use kde2d to visualize data on a sphere given in spherical
coordinates. Now the problem is, that phi == 2*pi = 0, so in principal
I have to connect (in a graphical view) the left and right border of my
plot (and the bottom and top). Has anyone any idea how to do that ?
Thanks,
I am moving this thread to R-devel, which is probably where it should have
started and it is getting even more technical.
On Thu, 13 Jan 2005, McGehee, Robert wrote:
Professor Ripley,
You are quite correct, I did have some S3/S4 confusion. Thank you. By
using an S3 method with the documented
On Fri, 14 Jan 2005, ecatchpole wrote:
I'm looking for something like Brian Ripley's glmmPQL that will handle
multinomial data. Does anyone know of anything?
It's a lot more complicated conceptually. A multinomial model has K-1
linear predictors which should probably have a correlated joint
On Fri, 14 Jan 2005, Manuel Metz wrote:
I want to use kde2d to visualize data on a sphere given in spherical
coordinates. Now the problem is, that phi == 2*pi = 0, so in principal I
have to connect (in a graphical view) the left and right border of my plot
(and the bottom and top). Has anyone
I have a large data matrix (4460X3500) and I want to sum row subsets in groups
of 10 (bin). Is there an efficient way to do this in R without using loops.
Looping takes forever to perform this task!
For example suppose we have the matrix
matrix(1:12,6,2)
[,1] [,2]
[1,]17
[2,]2
An additional problem separate from it not recognizing the figure
environment is that I guess you ran pdflatex (to get a .pdf file and not
a .dvi) and this indeed accepts only files in the formats below.
Try:
pdf(file=c:/myfigure.pdf)
plot(sin(1:10), pch=+)
dev.off()
Xianggui QU wrote:
Hi,
What you need is matrix multiplication:
rbind( c(1,1,0,0,0,0),
c(0,0,1,1,0,0),
c(0,0,0,0,1,1) ) %*% M
Bendix
--
Bendix Carstensen
Senior Statistician
Steno Diabetes Center
Niels Steensens Vej 2
DK-2820 Gentofte
Denmark
tel: +45 44 43 87 38
mob: +45 30 75 87 38
I intend to port an R project from Linux to Windows.
It involves C code that is loaded via dyn.load().
I could manage to produce a 'dll' File using cygwin which seems to be
o.k.
Now, using dyn.load(pcr.dll) i get:
Error in dyn.load(x, as.logical(local), as.logical(now)) :
unable to load
Hi Stephen,
you could try something like this:
mat - matrix(rnorm(4460*3500), 4460, 3500)
###
d - dim(mat)
bin - 10
system.time(out - lapply(split(mat,rep(seq(1, d[1]/bin), each=bin)),
function(x){dim(x) - c(bin, d[2]); colSums(x)}), gcFirst=TRUE)
out - matrix(unlist(out,
Here's a simple and efficient way
A - matrix(1:12,6,2)
nbin - 2
dim(A) - c(nbin, nrow(A)/nbin, 2)
B - colSums(A)
dim(B) - dim(A)[-1]
B
[,1] [,2]
[1,]3 15
[2,]7 19
[3,] 11 23
On Fri, 14 Jan 2005, Stephen Nyangoma wrote:
I have a large data matrix (4460X3500) and I want to sum
I fitted a GAM model with Poisson distribution to a data with about 200
observations. I noticed that the plot of the residuals versus fitted values
show a trend. Residuals tend to be lower for higher fitted values. Because,
I'm dealing with count data, I'm thinking that this might be due to
Tiago == Tiago R Magalhaes [EMAIL PROTECTED]
on Mon, 29 Nov 2004 16:41:30 + writes:
Tiago HI
Tiago the scan function when only one item is read says:
Tiago scan()
Tiago 1: 3.3
Tiago 2:
Tiago Read 1 items
Tiago [1] 3.3
Tiago I hope my english is not
[EMAIL PROTECTED] writes:
I intend to port an R project from Linux to Windows.
It involves C code that is loaded via dyn.load().
I could manage to produce a 'dll' File using cygwin which seems to be
o.k.
Now, using dyn.load(pcr.dll) i get:
Error in dyn.load(x, as.logical(local),
Hello all!
System: R2.0.1, W2k.
All packages apdated with update.packages().
I use the default graphic device for my system (no active choice).
I´m trying to fine-tune scatterplot() graphs (package car), for
publication.
I want to control plotting symbols and colours to match plots from
other,
Hi
This is a difference between the way matrices and data frames work I
guess. I want to replace the NA values in a data frame by 0, and the
code works as long as the data frame in question actually includes an NA
value. If it doesn't, there is an error:
df -
I am moving to using R more and more as a computational platform and
I use Microsoft Excel at the moment in exactly the way you describe in
WRITING ASCII FILES as a staging area for data, to gather data and make
simple calculations. I've have experienced problems with saving to .csv
files, and
2nd Workshop Ensemble Methods
Max Planck Institute Tuebingen, Germany
March 4-5, 2005
The second workshop on ensemble methods will take place at the Max
Planck Institute Tuebingen (Germany) on March 4-5, 2005. This workshop
is jointly organised by the German working groups Computational
Statistics
michael watson (IAH-C) wrote:
Any help would be appreciated. I could just convert the data frame to a
matrix, execute the code, then convert it back to a data frame, but that
appears long winded.
Slightly less long-winded (but probably a worse solution than some
R-guru is about to give you)
Hi Mick,
try the following:
dat[] - lapply(dat, function(x) ifelse(is.na(x), 0, x))
dat
I hope it helps.
Best,
Dimitris
Dimitris Rizopoulos
Ph.D. Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven
Address: Kapucijnenvoer 35, Leuven, Belgium
Tel: +32/16/336899
On Fri, 14 Jan 2005, michael watson (IAH-C) wrote:
Hi
This is a difference between the way matrices and data frames work I
guess. I want to replace the NA values in a data frame by 0, and the
code works as long as the data frame in question actually includes an NA
value. If it doesn't, there is
Thank you for the answers. Checking for NAs using any() will help.
I just thought it was worth mentioning because the behaviour of data
frames (which throw an error) is different to the behaviour of matrices
(which don't), and that might not be expected.
mat - matrix(c(1,1,1,1),nrow=2,ncol=2)
Mick,
The actual error is telling:
df - data.frame(c1=c(1,1,1),c2=c(2,2,2))
df[is.na(df)] - 0
Error in [-.data.frame(`*tmp*`, is.na(df), value = 0) :
rhs is the wrong length for indexing by a logical matrix
If you look at is.na(df), you will see that it is all FALSE, of course.
The
Thanks everybody for their answers!
Cheers,
fede
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Hello
I am new to R and I cannot overcome the 2 problems
1.Defining my own color scale for a value on a scatterplot
2.Adding legend showing values for colours
I have a set of points in area of my research with coordinates X and Y and
a value which I would like to show with colors.
#values for
I think Dirk's recommendation is better. If the system lacks tools to
compile R from source, install.packages() will not work for many packages,
as that compile packages from source.
Andy
ps: Thanks to those who pointed out about availability of dev tools in
Xandros Unbuntu. I still can't
Hi Andy,
I certainly deffer to both you and Dirk on this ... my suggestions were only
ment to get an installation of R up and running on Xandros. I'm by no means
an expert, as you can see by my missuse of the term 'dependencies'. I'd had
trouble using apt-get so I fiddled with Xandros Networks
Hi Mohamed,
I certainly deffer to both you and Dirk on this ... my
suggestions were only
ment to get an installation of R up and running on Xandros.
If you are using Xandros, then your information is better than my
speculation. If you were able to use install.packages(), you must have the
Your solution suggests that it works as expected on single columns
so in the special case that one is really only interested in NAs in one
particular column then the original can be made to work, e.g. for
column c2:
DF$c2[ is.na(DF$c2) ] - 0
Dimitris Rizopoulos dimitris.rizopoulos at
Henrik Andersson wrote:
I am moving to using R more and more as a computational platform and
I use Microsoft Excel at the moment in exactly the way you describe in
WRITING ASCII FILES as a staging area for data, to gather data and
make simple calculations. I've have experienced problems with
I get around the dependence on how things are displayed in Excel by
using
VBA to write my data files. Then you get full precision and you can use
white space as a separator if you like.
Sub WriteRangeToFile(rData as Excel.Range, fName as String)
Application.ScreenUpdating = False
On Fri, 2005-01-14 at 12:37 +0100, Henrik Andersson wrote:
I am moving to using R more and more as a computational platform and
I use Microsoft Excel at the moment in exactly the way you describe in
WRITING ASCII FILES as a staging area for data, to gather data and make
simple calculations.
Does someone know the meaning of the following values relating to the
cca-command?
iner, inercum, inerC, inercumC
Thanks!
DOMI
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
Dear list,
there are now 400 packages available on CRAN.
Would it be useful to classify these packages
according to what they do (e.g. classification,
graphics, spatial statistics), to assist the user in
finding the appropriate package for their problem? Or
perhaps the search facility is enough.
Dear CG,
When I wrote the scatterplot() function I expected it to be used in data
exploration and didn't anticipate its use for publication graphics.
That said, it wouldn't be hard to add the full set of cex* arguments and to
pass these through to plot() and points(). I'll do this for the next
dan,
good idea. i propose an additional field called 'keywords' to the package
source file 'description'. this would be better than a single taxonomy of
packages.
Robert Sams
SANCTUM FI LLP
email: [EMAIL PROTECTED]
Authorised and Regulated by the FSA.
Sending encrypted mail:
See
I'm trying to overlay a density plot on a previously plotted histogram.
However, i need to use the same axis as of the 1st(histogram) plot to plot the
second. My second plot is creating its own axis and causing my plot to extend
the entire histogram instead of getting a subplot on a portion of
hi,
I would like to subsample the array c(1:200) at random into ten subsamples
v1,v2,...,v10.
I tried with to go progressively like this:
x-c(1:200)
v1-sample(x,20)
y-x[-v1]
v2-sample(y,20)
and then I want to do:
x-y[-v2]
Error: subscript out of bounds.
I searched CRAN for material on gant (or Gantt) charts, those schedule
plots that look like horizontal barplots where the x-axis is time.
Finding none, I wrote the following and am submitting it to the archive
so that it may help someone in the future.
plotfile - gant.eps
regular.color -
Henrik Andersson h.andersson at nioo.knaw.nl writes:
:
: I am moving to using R more and more as a computational platform and
: I use Microsoft Excel at the moment in exactly the way you describe in
: WRITING ASCII FILES as a staging area for data, to gather data and make
: simple
Hi All,
In package e1071 for SVM based classification, one can get a probability
measure for each prediction. I like to know what is method that is used for
calculating this probability. Is it calculated using logistic link function?
Thanks for your help.
Regards,
Raj
[EMAIL PROTECTED] wrote:
hi,
I would like to subsample the array c(1:200) at random into ten subsamples
v1,v2,...,v10.
I tried with to go progressively like this:
x-c(1:200)
v1-sample(x,20)
y-x[-v1]
v2-sample(y,20)
and then I want to do:
x-y[-v2]
Error: subscript out of bounds.
Let's do a
Jagarlamudi, Choudary wrote:
I'm trying to overlay a density plot on a previously plotted histogram. However, i need to use the same axis as of the 1st(histogram) plot to plot the second. My second plot is creating its own axis and causing my plot to extend the entire histogram instead of getting
From: [EMAIL PROTECTED]
hi,
I would like to subsample the array c(1:200) at random into
ten subsamples
v1,v2,...,v10.
I tried with to go progressively like this:
x-c(1:200)
v1-sample(x,20)
y-x[-v1]
v2-sample(y,20)
This only worked because your original data happens to be
[EMAIL PROTECTED] a écrit :
hi,
I would like to subsample the array c(1:200) at random into ten subsamples
v1,v2,...,v10.
I tried with to go progressively like this:
x-c(1:200)
v1-sample(x,20)
y-x[-v1]
v2-sample(y,20)
and then I want to do:
x-y[-v2]
Error: subscript out of
[EMAIL PROTECTED] writes:
hi,
I would like to subsample the array c(1:200) at random into ten subsamples
v1,v2,...,v10.
(Why are you c()'ing a single vector?)
I tried with to go progressively like this:
x-c(1:200)
v1-sample(x,20)
y-x[-v1]
v2-sample(y,20)
and then I want to
On Fri, 2005-01-14 at 16:28 +, Gabor Grothendieck wrote:
Henrik Andersson h.andersson at nioo.knaw.nl writes:
:
: I am moving to using R more and more as a computational platform and
: I use Microsoft Excel at the moment in exactly the way you describe in
: WRITING ASCII FILES as a
[EMAIL PROTECTED] wrote:
Hello
I am new to R and I cannot overcome the 2 problems
1.Defining my own color scale for a value on a scatterplot
2.Adding legend showing values for colours
I have a set of points in area of my research with coordinates X and Y and
a value which I would like to show with
Hello 'R' Users,
I am very new on 'R', so excuse me if I ask something wrong.
I have ASCII data and the colums of the data are looks like :-
!-
time,yr,mo,dy,hr,min,sec,lat,lon,ht,co2obs,sigma,co2model
--
-
--
!
Each column has data value. Now I
On Fri, 14 Jan 2005, Dan Bebber wrote:
Dear list,
there are now 400 packages available on CRAN.
Would it be useful to classify these packages
according to what they do (e.g. classification,
graphics, spatial statistics), to assist the user in
finding the appropriate package for their
Is it possible to get the empirical (sandwich) S.E. estimates for the
fixed effects in lme () (thus allowing possibly correlated errors within
the group)? In SAS you can get it by the 'empirical' option to PROC MIXED.
Cheers,
Michael
--
Na (Michael) Li, Ph.D.
Division of
I worry about this proposal because of the following example.
One of the packages I use a lot (because it solves a problem I've
been mulling for 10 years about how to test individual subjects)
is multtest. That is part of Bioconductor, and the original
purpose of it was (apparently) for DNA
Cut and paste the following function (to be in next release of R, I believe)
and type:
RSiteSearch(election maps)
This, of course, assumes that you have Internet connection.
Andy
RSiteSearch - function(string,
restrict=c(Rhelp02a,Rhelp01,functions,docs),
On Fri, 14 Jan 2005, Jonathan Baron wrote:
I worry about this proposal because of the following example.
I don't think you will get something that satisfies everyone in all
cases...and not doing it for the reason will result in still having 400
unsorted packages on CRAN.
One of the packages I
Weiwei Shi writes:
Hi,
I happen to re-write my codes to save memory and my
approach is write my obj into file first and later I
load it.
However, it seems like:
load(filename) can load the object but the function
returns the name of the object instead of the
reference to it. For example,
???
correlated within group errors are explicitly modeled by corStruct classes.
See ?lme and Chapter 5.3 in Bates and Pinheiro.
-- Bert Gunter
Genentech Non-Clinical Statistics
South San Francisco, CA
The business of the statistician is to catalyze the scientific learning
process. - George E.
Thank you very much for your time and response.
Idid the following:
x1-range(xvals)
hist(xvals,xlim=x1,ylim=NULL)
#now i'd like to plot a density plot only for the 1st 2 bars of the histogram
by passing xvals in that range.
ds-density(xvals[1:fr],bw=30,n=fr)
plot(ds$y ~ xvals[1:fr])
My second
Right, but you still can sandwich them if you want.
(I recently did that in Proc MIXED, but Michael, I'm not sure how to
do it using lme).
best,
-tony
On Fri, 14 Jan 2005 11:16:10 -0800, Berton Gunter
[EMAIL PROTECTED] wrote:
???
correlated within group errors are explicitly modeled by
A.J. Rossini wrote:
Right, but you still can sandwich them if you want.
(I recently did that in Proc MIXED, but Michael, I'm not sure how to
do it using lme).
best,
-tony
The sandwich estimator can help if the model is misspecified but at a
cost of worse precision in estimating variances.
But similar to Jon's issue, one might imagine the following future
scenario, in 10 years, after R dominates statistics:
but it's not in R's biostat task list, so I don't have to know about it!
(I am NOT being serious, but I've heard similar justifications in the
past, i.e. it's not part of
Hi Dominique,
please, if you have some questions about a particular package, do not send
to R-Help. There is a list about ade4, see details at:
http://pbil.univ-lyon1.fr/ADE-4/adelist.html
cca is a correspondence analysis with constraint imposed by external
information. Redundancy analysis
Hello,
I'd like to remove color bar from filled.contour
plot.
Is it possible and how?
I also want to overlay contour plot on
filled.contour
but due to scaling of the filled.contour to
account for the color bar, aspect ratio for contour
is different from filled.contour. Can this problem
be solved?
I thought I would take a look at the StatDataML package for some work
I'm doing, but I receive the following error when starting:
version
_
platform i386-pc-mingw32
arch i386
os mingw32
system i386, mingw32
status
major
I work in biology and I use mixed-model for my data analysis
In a scientific paper, the author wrote:
All continuous exploratory variables were centred on their median value
prior to inclusion in the analysis (Pinheiro Bates, 2000).
They refer to the book Mixed-effects model in S and S-Plus by
You might want to try ?image like in :
require(MASS)
x - rnorm(50)
y - rnorm(50)
d - kde2d(x,y)
image(d,col=terrain.colors(50))
contour(d,add=T)
Hope this is what you are looking for.
Romain.
m p a écrit :
Hello,
I'd like to remove color bar from filled.contour
plot.
Is it possible and how?
I also
On Fri, 2005-01-14 at 15:52 -0600, Shawn Way wrote:
I thought I would take a look at the StatDataML package for some work
I'm doing, but I receive the following error when starting:
version
_
platform i386-pc-mingw32
arch i386
os mingw32
Thanks, but this does not work correctly.
Contour lines with contour do not coincide
with those produced by image. Also image plots
contour lines with right angles while contour plots
contour lines taking the diagonal between neighboring
points. All looks quite ugly.
If there are options to make
(Posted because private reply to sender was kicked back).
Note that you said centered **response** variables in your original
original post, not centered **explanatory** variables, which is what (your
misquote of) Bates and Pinheiro said. ** Big Difference!** If you don't know
what the difference
Hello again,
look at the data provided by d, it's a list with:
- x : the 1st coordinate
- y : the 2nde coordinate
- z : a matrix of dimension 25 x 25 (the rectangles shown by image are
the representation of those 625 points, the contour
function uses interpolation.
See the n parameter in kde2d
Hello, it would be nice (as it is asked in the posting guide) to post
something i can actually run on my machine, since i have no idea what
are xvals, fr, .. i can only assume what you want.
1) if you want the second plot to overlay the first one (generated by
hist) you might want to use
Yes, this works perfectly. I still need to
change the labeling on the label bar since it
labels reals (1, 1.5, 2., 2.5 etc) which does not
make sense for this plot - it's a number of iterations
which needs to integer. Anything can be done?
Thanks,
Mark
--- Thomas Lumley [EMAIL PROTECTED] wrote:
Noob alert
I've been through most of the available documentation at a fairly high
level - if I've missed the obvious, just point me in the right
direction.
What I'd like to build:
Windows XP system (actually, I'd prever a *nix but must use Win).
Backend data stored in MySQL db
Front end user
Hi Andy,
Your comment that the difference sounded peculiar got me checking my code
and notes again, especially since all OS installations were in fact on the
same machine.
I'm embarrassed to say that discovered I had added an extra few loops to my
algorithm when I ran it on Xandros ...
On Fri, 14 Jan 2005, m p wrote:
Yes, this works perfectly. I still need to
change the labeling on the label bar since it
labels reals (1, 1.5, 2., 2.5 etc) which does not
make sense for this plot - it's a number of iterations
which needs to integer. Anything can be done?
You can specify the
On 1/14/05 19:34, Donna-n-Doug Finner [EMAIL PROTECTED] wrote:
Noob alert
I've been through most of the available documentation at a fairly high
level - if I've missed the obvious, just point me in the right
direction.
What I'd like to build:
Windows XP system (actually, I'd prever a
f(x) = .01*(1.2^y) integrated from 0 to x
Find x such that
f(x) = some number, say 4
How do we do it in R?
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
Greetings,
I am working on a stylometric analysis of some latin texts; one of the
latest stylometric techniques involves using principal components
analysis. Not being a statistician, I can't really fully rely on PCA
as my primary tool, since I don't really understand the statistics
aar paar aar420 at hotmail.com writes:
:
: f(x) = .01*(1.2^y) integrated from 0 to x
:
: Find x such that
:
: f(x) = some number, say 4
:
: How do we do it in R?
You don't need R for this.
Your integral can be solved analytically, check any
table of integrals, and the rest can be solved
I am encountering an font error when using either plot() or pairs() for a
scatter plot matrice under some circumstances. For instances pairs(hills)
using the hills data set in MASS results in the following error:
Error in text.default(x, y, txt, cex = cex, font = font) :
X11 font at
This cropped up in a previous posting here (Monday this week, in fact),
responded to by Prof. Ripley. It also cropped up on the Bioconductor mailing
list before Christmas, in relation to the AnnBuilder package. For those who
have reposTools installed, the solution provided by John Zhang was, from
On Friday 14 January 2005 23:29, John Dougherty wrote:
I am encountering an font error when using either plot() or pairs() for a
scatter plot matrice under some circumstances. For instances pairs(hills)
using the hills data set in MASS results in the following error:
Error in text.default(x,
81 matches
Mail list logo
