The 'r' in rlm is for 'robust', so it does not compute a residual sum of
squares (which is not robust), but rather a robust estimate of the scale.
That *is* what the help page ?summary.rlm says:
sigma: The scale estimate.
stddev: A scale estimate used for the standard errors.
As to
Remove all the blank spaces in your package's name and try again.
Uwe Ligges
Dominik Locher - bondsearch.ch wrote:
Hi
I'm trying to compile R packages on my vista (I'm very new on vista...). I
installed Rtools, RExcel on a specific folder: c:\Rtools and gave full
access. Unfortunately,
Patrick Connolly wrote:
On Mon, 08-Dec-2008 at 02:05AM +0800, Berwin A Turlach wrote:
| G'day Wacek,
|
| On Sat, 06 Dec 2008 10:49:24 +0100
| Wacek Kusnierczyk [EMAIL PROTECTED] wrote:
|
| []
| there is, in principle, no problem in having a high-level language
| perform the
Berwin A Turlach wrote:
I am not surprised about CS guys never learning about these issues. As
long as you play around with data bases (their organisation c),
sorting algorithms, artificial intelligence (at least when I attended a
lecture on this) you do not need to know about these issues.
I would like to estimate a 95% highest density area for a multivariate
parameter space (In the context of anova). Unfortunately I have only
experience with univariate kernel density estimation, which is remarkebly
easier :)
Using Gibbs, i have sampled from a posterior distirbution of an Anova
Hello Bernd,
by definition, a VAR does only include **lagged endogenous** variables.
You might want consider SVAR() contained in the same package, or fit a
VECM (see CRAN package 'urca').
Best,
Bernhard
Hi useRs,
Been estimating a VAR with two variables, using VAR() of the
package vars.
I all,
is there a way with R to perform an exact permutation test to replace the
wilcoxon test to compare paired series and/or to perform pairwise multiple
comparisons for related series after a Friedman test ?
Thanks
Gilles
__
R-help@r-project.org
Csima Gabriella wrote:
Dear Everyone,
I wrote a Taylor diagram program in R, but it was not very general, so I was
happy to find the Taylor.diagram in the plotrix package.
On the other hand I can find many problems with the pos.cor=TRUE case, in other
words, when we use only the first quarter
Hello Kamlesh,
have a look at: fUnitRoots, tseries, urca, uroot
Best,
Bernhard
Dear sir,
I am a new user of R statistical package. I want to perform
adf.test(augmented dickey fuller test), which packages I need
to install in
order to perform it. I am getting following message on my
Thank you for your help!
Sorry, for bothering you again..
I still have trouble combining within and between subject factors.
Interactions of within factors and D having only 2 levels work well.
How can I get the main effect of D? I have tried anova(mlmfitD, mlmfit).
With D having 3 levels I
Hi Gabriela,
as suggested by Jim I'll have a look as soon as possible (but only
tomorrow I'm afraid).
Maybe you could provide me with a dataset of yours which shows the
problem (use direct postin gt o my own address
[EMAIL PROTECTED], instead of posting the data to Rlist) ?
Thanks,
Olivier
G'day Wacek,
On Mon, 08 Dec 2008 13:13:33 +0100
Wacek Kusnierczyk [EMAIL PROTECTED] wrote:
Berwin A Turlach wrote:
I am not surprised about CS guys never learning about these
issues. As long as you play around with data bases (their
organisation c), sorting algorithms, artificial
Thank you very much for your help! That code does exactly what I was looking
for ( I don't have experience with lattice yet )
Ciao,
Antje
[EMAIL PROTECTED] schrieb:
layout(matrix(c(1,2,3,4), nrow = 2, byrow = TRUE))
plot(rnorm(100))
plot(rnorm(200))
plot(rnorm(300))
plot(rnorm(400))
Now,
G'day Wacek,
On Sun, 07 Dec 2008 21:09:36 +0100
Wacek Kusnierczyk [EMAIL PROTECTED] wrote:
g'evening.
Not the done thing.
c'mon, a person from central europe can't possibly be unaware of this
joke.
I wouldn't call Norway central Europe, but then I also guess that you
are not really
Anamika Chaudhuri [EMAIL PROTECTED] 08/12/2008 03:46:34
I am using a random intercept model with SITEID as random and NAUSEA
as
outcome.
Hardly surprising ;-)
More seriously, your data set fragment had only one level for SITEID. I
assume there were actually more levels?
S
Dear Everyone,
I wrote a Taylor diagram program in R, but it was not very general, so I was
happy to find the Taylor.diagram in the plotrix package.
On the other hand I can find many problems with the pos.cor=TRUE case, in other
words, when we use only the first quarter of the space (positive
Here's a paired-samples permutation test. Note that this does not obtain a
p-value based on all possible randomization orders, a test that by some
nomenclatures would be called a Randomization test and which I presume is
what you desire given the use of the word exact in your query.
Instead, the
Ben Bolker's response to a glmmPQL question below raises a question.
Does the issue of bias with binomial data reported by Breslow (2003)
remain valid with respect specifically to Ripley's treatment of PQL in
glmmPQL? Breslow makes no reference to this particular implementation.
He does discuss
Dear all,
I'm a newbie in R.
I have a 45x2x2x8 design.
A dataframe stores the metadata of trials. And each trial has its own
data file: I used read.table to import every trial into R as a
dataframe (variable).
Now I dynamically ask R to retrieve trials that fit certain selection
criteria, so I
Hi,
I'm not sure that I fully understand your problem, so forgive me if my
suggestion is irrelevant :
maybe you should have a look to function geod.dist , in the oce package.
Hope this helps. Olivier
porzycka wrote:
Hello,
I have been worked with R for 3 weeks. I have one problem. I could
Perhaps you will find useful code in the examples for freidman_test
within package coin. They offer an implementation of the Wilcoxon-
Nemenyi-McDonald-Thompson test.
--
David Winsemius
On Dec 8, 2008, at 6:15 AM, LE PAPE Gilles wrote:
I all,
is there a way with R to perform an exact
hello all,
where I find a example or tutorial of RGtk2 package?
I would like to know about the gladeXML functions in R.
thanks in advance
Cleber Borges
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read
On Dec 8, 2008, at 10:11 AM, tsunhin wong wrote:
Dear all,
I'm a newbie in R.
I have a 45x2x2x8 design.
A dataframe stores the metadata of trials. And each trial has its own
data file: I used read.table to import every trial into R as a
dataframe (variable).
Now I dynamically ask R to
Dear:
I try to analysis multinomial logistic regression using vglm in VGAM package.
However, I wonder how many levels of responses variable this command is
suitable. I saw the examples in google search works for 3 levels, say 1,2,3.
However, my response variable is more than 3 levels. Is it
Dear ALL:
Could you please eamil me how to simulate Mixed Seasonal ARMA (p,q)x(P,Q)12
models [say ARMA(0,1)x(1,0)12 ]from R.
With many thanks.
Abou
==
AbouEl-Makarim Aboueissa, Ph.D.
Assistant Professor of Statistics
Department of Mathematics Statistics
?get
Sent from my iPhone
On Dec 8, 2008, at 7:11, tsunhin wong [EMAIL PROTECTED] wrote:
Dear all,
I'm a newbie in R.
I have a 45x2x2x8 design.
A dataframe stores the metadata of trials. And each trial has its own
data file: I used read.table to import every trial into R as a
dataframe
Hi,
how can I save an image as a metafile?
I know within windows you can do a right click and then 'save image as
metafile'
but I use Mac OS X...I know as well that mac users have a right click as
well, but
it does not work.
Is there a command in R for saving images as metafiles?
Regards,
Hello everyone,
Â
I am a very new user of R and I have a query about the cuminc function in the
package cmprsk. In particular I would like to verify that I am interpreting the
output correctly when we have a stratification variable.
Â
Hypothetical example:
Â
group : fair hair, dark hair
Dear R Users,
Is there a package or some functionality in R which returns statistics on
runs of numbers, i.e. series of numbers with similar qualities in a time
series ? For example, the number of +ves,-ves, histograms on cumulations
in runs, etc. ?
Thanks in advance,
Tolga
Generally, this
Dear Tsunhin,
Take a look at ?get
HTH,
Jorge
On Mon, Dec 8, 2008 at 10:11 AM, tsunhin wong [EMAIL PROTECTED] wrote:
Dear all,
I'm a newbie in R.
I have a 45x2x2x8 design.
A dataframe stores the metadata of trials. And each trial has its own
data file: I used read.table to import every
Thanks Jim and All!
It works:
tmptrial - trialcompute(trialextract(
get(paste(t,tmptrialinfo[1,2],tmptrialinfo[1,16],.gz,sep=)) ,
tmptrialinfo[1,32],secs,sdm),secs,binsize)
Can I use assign instead? How should it be coded then?
Thanks!
- John
On Mon, Dec 8, 2008 at 10:40 AM, Jim Holtman
I suggested the use of grconvertY if you want to place something exactly n
inches from the bottom, then grconvertY will give the user (or other)
coordinates that match, or if you want to place something exactly 10% of the
total device height from the bottom, etc.
What to do in replotting a
Many thanks,
Tolga
Albyn Jones [EMAIL PROTECTED]
08/12/2008 17:16
To
[EMAIL PROTECTED]
cc
r-help@r-project.org
Subject
Re: [R] statistics on runs of numbers
rle(x) gives the run length encoding of x.
rle(x0) or rle(sign(x)) will do this for positive and negative values of
x.
albyn
This is to announce that we plan to release R version 2.8.1 on Monday,
December 22, 2008.
Release procedures start Friday, December 12.
Those directly involved should review the generic schedule at
http://developer.r-project.org/release-checklist.html
The source tarballs will be made
Hej!
I have the following problem:
I would like to do partial correlations on non-parametric data. I checked
pcor (Computes the partial correlation between two variables given a set
of other variables) but I do not know how to change to a Spearman Rank
Correlation method
Hi Jens,
I assume you are using package dlm.
To add covariates to the system equation the workaround is to include
psi in the state vector. Suppose your observation and system equations are
as follow:
Y[t] = F[t]theta[t] + v[t], v[t] ~ N[0,V] #observation equation
theta[t] = G[t]theta[t-1] +
I'm not sure if the gee package is still actively maintained, but I for one
find it extremely useful. However, I've come across a few infelicities that I'm
hoping could be resolved for future versions. Hope it's okay to list them all
in one post! They are:
(1) AR(1) models don't fit when
Dear experts
I need to fit a model with nested random effects (cow within herd and season),
and Im not sure how to do it with R.
Using LME I know how to fit: cow within herd
fit4 - lme(milk ~ days + season + season*days , data=xxx, random = ~1 + days|
herd/cow)
How I can fit: cow
Skotara wrote:
Thank you for your help!
Sorry, for bothering you again..
I still have trouble combining within and between subject factors.
Interactions of within factors and D having only 2 levels work well.
How can I get the main effect of D? I have tried anova(mlmfitD, mlmfit).
With D
Hello all,
I wish to plot several hundred data groups in a boxplot(), with
sensible labels on the x axis. For small datasets, this is possible by
using the names parameter to the boxplot. However, for several
hundred boxplots, boxplot() displays a tick on the x axis for every one.
For my
Sorry, I will be more precise. Here is an example (simplified) of graph
I want :
8
symboles - c(3,4,5,6)
dn - rbind(matrix(rnorm(20),5))
layout(matrix(c(1,1,1,2,2,3),3))
for(i in 1:3)
matplot(dn,type=b,xlab=+: a x: b ???: c ???: d,pch=symboles)
8
But instead of ???, I
In the long run it will probably make your life much easier to read all the
dataframes into one large list (and have the names of the elements be what your
currently name the dataframes), then you can just use regular list indexing
(using [[]] rather than $ in most cases) instead of having to
Try this:
symboles - c(3,4,5,6)
dn - rbind(matrix(rnorm(20),5))
layout(matrix(c(1,1,1,2,2,3),3))
for(i in 1:3){
matplot(dn,type=b,xlab=,pch=symboles)
legend(grconvertX(0.5,'nfc'), grconvertY(0,'nfc'),
xjust=0.5, yjust=0,
pch = unique(symboles),
legend = c(a,b,c,d),
Dear R ExpeRts,
I'm having memory difficulties using mahalanobis distance to trying to cluster
in R. I was wondering if anyone has done it with a matrix of 6525x17 (or
something similar to that size). I have a matrix of 6525 genes and 17 samples.
I have my R memory increased to the max and
A good example of using glade with RGtk2 is the rattle package.
See: http://rattle.togaware.com/
On Mon, Dec 8, 2008 at 8:15 AM, Cleber Nogueira Borges
[EMAIL PROTECTED]wrote:
hello all,
where I find a example or tutorial of RGtk2 package?
I would like to know about the gladeXML functions
I think actually it's both lazy evaluation and scoping. Here is how I
understand it.
Consider:
flist - vector(list,2) ## creates the empty list
for(i in 1:2)slist[[i]] - function()i
Now the RHS of the assignment is a function that returns the value of i.
That is:
flist
function()i
not sure which library 'pcor' is in, but why don't you just use the ranks of
the variables and then perform the correlation on the ranks:
x-sample(1:10,10,rep=T)
y-x+ sample(1:10,10,rep=T)
cor(x,y)
cor(rank(x),rank(y))
HTH
david freedman
Jürg Brendan Logue wrote:
Hej!
I have the
On Dec 8, 2008, at 7:51 AM, mentor_ wrote:
Hi,
how can I save an image as a metafile?
I know within windows you can do a right click and then 'save image as
metafile'
but I use Mac OS X...
Which means this question would be better posed on the Mac OS list.
I know as well that mac users
I want to combine all dataframes into one large list too...
But each dataframe is a 35 columns x varying number of rows structure
(from 2000 to 9000 rows)
I have ~1500 dataframes of these to process, and that add up to
1.5Gb of data...
Combining dataframes into a single one require me to
On Mon, 8 Dec 2008, David Winsemius wrote:
On Dec 8, 2008, at 7:51 AM, mentor_ wrote:
Hi,
how can I save an image as a metafile?
I know within windows you can do a right click and then 'save image as
metafile'
but I use Mac OS X...
Which means this question would be better posed on the
I really don't understand your concern. Something like:
nms - c('file1','file2','file3')
my.data - list()
for (i in nms) my.data[[ i ]] - read.table(i)
Will read in the files listed in the nms vector and put them into the list
my.data (each data frame is a single element of the list). This
Thank you very much,
That's very close to what I want. Well, I would have love to let the
user manualy resize the graph, but I am not sur it is possible. So
that's enouth for me, thanks a lot.
Christophe
PS : with grconvertX(0.5,'nfc'), the legend is not center.
With grconvertX(0.535,'nfc'),
Dear R-users,
I have found on the list several posts addressing the issue of getting data for
a list of defined files. However, I found nothing about scanning a given folder
and retrieving the list of, let's say, .txt files (I probably used the wrong
keywords). Could someone tell me which posts
Sebastien,
see ?list.files, especially the 'pattern' argument. For your
particular case it is
list.files(pattern=\\.txt$)
You might want to use the 'ignore.case' argument as well.
Gabor
On Mon, Dec 8, 2008 at 10:53 PM, [EMAIL PROTECTED] wrote:
Dear R-users,
I have found on the list
?files
and the links therein. Seems like a most obvious keyword to me for asking
about files ...
Cheers,
Bert Gunter
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf Of [EMAIL PROTECTED]
Sent: Monday, December 08, 2008 1:54 PM
To: r-help@r-project.org
Possible, yes (see fortune('Yoda')), but doing it can be a bit difficult.
Here is one approximation using an independent normal kernel in 2 dimensions:
kde.contour - function(x,y=NULL, conf=0.95, xdiv=100, ydiv=100, kernel.sd ) {
xy - xy.coords(x,y)
xr - range(xy$x)
yr -
Dear all,
For some time now, R is becomming more and more popular in more and
more countries. France is for sure one of them, but french people
being french one of the obstacle they might tackle is the lack of
documentation and support in their native language.
To offer this support in french an
On Dec 8, 2008, at 4:02 PM, Prof Brian Ripley wrote:
On Mon, 8 Dec 2008, David Winsemius wrote:
To find out what devices are available, try:
capabilities()
jpeg png tifftcltk X11 aqua http/ftp
sockets libxml fifo clediticonv NLS
TRUE TRUE
I've read in many places that R semantics are based on Scheme semantics. As
a long-time Lisp user and implementor, I've tried to make this more precise,
and this is what I've found so far. I've excluded trivial things that
aren't basic semantic issues: support for arbitrary-precision integers;
Dear Peter and Nils,
I hesitate to repeat this (though I'm going to do it anyway!), but it's
quite simple to get these tests from Anova() in the car package. Here's an
example from ?Anova of a repeated-measures ANOVA with two within and two
between-subject factors:
-- snip
On Mon, 8 Dec 2008, Stavros Macrakis wrote:
I've read in many places that R semantics are based on Scheme semantics. As
a long-time Lisp user and implementor, I've tried to make this more precise,
and this is what I've found so far. I've excluded trivial things that
aren't basic semantic
Dear R-users,
I'm currently using the multcomp package to produce plots of means with 95%
confidence intervals
i.e.
mult-glht(lm(response~treatment, data=statdata),
linfct=mcp(treatment=Means))
plot(confint(mult,calpha = sig))
Unfortunately the y-axis on the plot appears to be fixed and hence
On Sun, Dec 7, 2008 at 9:37 AM, Jon Loehrke [EMAIL PROTECTED] wrote:
Hi,
I am having an issue with a custom panel for lattice. The problem
comes when I try passing a groups argument.
Here is the custom panel, a wrapper for smooth spline. I copied
panel.loess and replaced the loess
On Sat, Dec 6, 2008 at 9:27 PM, Giam Xingli [EMAIL PROTECTED] wrote:
Hello everyone,
This is my first post to the mailing list, so I hope I am posting my message
the correct way.
I am trying to present my dataset in a 3d scatterplot using cloud() in the
{lattice} package. I hope to
See ?par and note the 'mar' parameter
Here's an example:
library(multcomp)
labs - c('short', 'medium', 'long')
treatment - gl(3, 10, labels = labs)
response - rnorm(30, mean=as.numeric(treatment))
mult - glht(lm(response ~ treatment),
linfct=mcp(treatment='Means'))
Hi all,
I want to draw n plots whose titles are Main effect of X_i, i= 1,
2, ..., n, respectively.
More specifically, I use:
par(mfrow = c(n/2, 2))
for ( i in 1:n)
{
plot(1:J, y[i, 1: J], main = 'Main effect of X_i')
}
This gives me a series of plots with the common title with a
I am a pretty new R user, I am running the latest linux version on xandros,
updated with some extra debian packages, and I also run the latest windows
version, but prefer linux.
I am having trouble downloading bayesm, it won't do it all from any of the
sites on the web, I resorted to this one,
Thanks Gabor
Gábor Csárdi a écrit :
Sebastien,
see ?list.files, especially the 'pattern' argument. For your
particular case it is
list.files(pattern=\\.txt$)
You might want to use the 'ignore.case' argument as well.
Gabor
On Mon, Dec 8, 2008 at 10:53 PM, [EMAIL PROTECTED] wrote:
see ?paste
main = paste (Main effect of , variable)
par(mfrow = c(n/2, 2))
for ( i in 1:n)
{
plot(1:J, y[i, 1: J], main = 'Main effect of X_i')
}
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do
Hi
Lorenzo Isella wrote:
Dear All,
I ma having a trouble in generating a figure containing 3 insets with
the gridBase package.
I always get an error message of the kind:
Error in gridPLT() : Figure region too small and/or viewport too large
No matter which parameters I choose. The
I am using a random intercept model with SITEID as random and NAUSEA as
outcome.
I tried using a dataset without missing values and changed my model
statement accordingly but still get the same error. Follwoing in an excerpt.
anal.data - read.table(nausea.txt, header=T, sep=\t)
Hi,
i'm trying to compile R on a Cray XT3 using pgi/7.2.1 - CNL (compute
node linux)
The R version is 2.8.0
this is the option
-enable-R-static-lib=yes
--disable-R-shlib
CPICFLAGS=fpic
FPICFLAGS=fpic
CXXPICFLAGS=fpic
SHLIB_LDFLAGS=shared
--with-x=no
SHLIB_CXXLDFLAGS=shared
Hello R users,
I would like to create several forestplots with the same X axis, so,
if you were to look at the plots lined up all the X axes would be
identical (and the different plots could be compared). Here is one
version of code I've used:
mytk10-c(0.1, 0.5, 1, 2, 5, 10)
pdf(file =
Hi,
Is there a way when doing logistic regression for the output to spit out
accuracy, sensitivity, and specificity?
I would like to know these basic measures for my model.
Thanks!
[[alternative HTML version deleted]]
__
R-help@r-project.org
I have date and time data which looks like this:
[,1] [,2]
[1,] 7/1/08 9:19
[2,] 7/1/08 9:58
[3,] 7/7/08 15:47
[4,] 7/8/08 10:03
[5,] 7/8/08 10:32
[6,] 7/8/08 15:22
[7,] 7/8/08 15:27
[8,] 7/8/08 15:40
[9,] 7/9/08 10:25
[10,] 7/9/08 10:27
I would like to use chron on it,
On Mon, Dec 8, 2008 at 11:52 PM, Tubin [EMAIL PROTECTED] wrote:
I have date and time data which looks like this:
[,1] [,2]
[1,] 7/1/08 9:19
[2,] 7/1/08 9:58
[3,] 7/7/08 15:47
[4,] 7/8/08 10:03
[5,] 7/8/08 10:32
[6,] 7/8/08 15:22
[7,] 7/8/08 15:27
[8,] 7/8/08 15:40
Hi experts of R,
Are there any functions in R to test a univariate series for long memory
effects, structural breaks and time reversability?
I've found for ARCH effects(ArchTest), for normal (Shapiro.test,
KS.test(comparing with randn) and lillie.test) but not for the above
mentioned.
Where can
Luke Tierney wrote:
R does not have macros.
Those are related -- because of lazy evaluation one does macros are
not needed to achive semantic goals (see for example tryCatch). Being
able to define friendlier syntax would sometimes be nice though (see
tryCatch again).
Also for some
I keep getting the following error when I look for minima in the series:
aa.peak - wavCWTPeaks (aa.tree)
Error in `row.names-.data.frame`(`*tmp*`, value = c(1, 0)) :
invalid 'row.names' length
How can I work it around ?
Thank you.
Regards,
Maura
Alice Messenger ;-) chatti anche con gli
Are the names of the rows the same as the time series that you are
using? I know that I am not being that helpful, but this seems like a
mismatch in the time series object. look at
length(rowname(your.data))
length(your.data[,1])
again it is always helpful to have reproducible code.
On Tue,
Please try the R-patched.version of R (see the posting guide or the FAQ).
That seems to be about an error in the R 2.8.0 sources that was corrected
in October, and only happens if you do not ask for X11 support (which most
R users need, including it seems all the pre-release testers).
On
thanks. i noticed a syntax error. a missing semi colon.
any idea about the error below?
cc -I. -I../../../src/include -I../../../src/include -I/usr/local/
include -DHAVE_CONFIG_H -fPIC -g -O2 -c Rsock.c -o Rsock.o
/opt/cray/xt-asyncpe/1.2/bin/cc: INFO: linux target is being used
cc -I.
82 matches
Mail list logo