Dear listers,
I am trying to get more familiar with concepts underlying generalised
linear mixed models, mainly through Venables and Ripley (fourth edition)
and the R-list archive. Of course, as a possibly tool-user biologist I
am not that easy with every détails of the mathematical aspects of
See the rw-FAQ Q2.6 for a more precise answer than has yet appeared in
this thread. A while back (for 2.1.0) some aspects of R were optimized
for usage from a slow (and possibly read-only) drive.
This does presume that the sites you visit will allow you to run programs
from an external drive,
On 11/16/05, Wassell, James T., Ph.D. <[EMAIL PROTECTED]> wrote:
> I am using the package nlme to fit a simple random effects (variance
> components model)
>
> with 3 parameters: overall mean (fixed effect), between subject
> variance (random) and within subject variance (random).
So to paraphra
On Wed, 16 Nov 2005, P Ehlers wrote:
> This should work:
>
> update.packages(ask = "graphics", repos = NULL,
> contriburl = "file:///g:/myFolder/myRepository"))
Only if the so-called repository contains (only) binary builds under R
2.2.x of packages for Windows.
It would be better to set up
Dear UseRs,
maybe is a silly question: how can I get Empirical CDF
values from an object created with ecdf()?? Using
print I obtain:
Empirical CDF
Call: ecdf(t)
x[1:57] =4.1,4.4,4.5, ..., 491.3,
671.27
Thanks in advance.
Regards,
Vito
Diventare costruttori di soluzioni
Became
Firstly, these are all very basic questions, thah you coukld probobly answer
yourself by searching the manual and help files. Nevertheless, here are the
answers to give you a little head start.
> Dear R expert,
> The following is my questions:
>
> 1. How to generate two sequences of 150 uniform
Hello,
I am interested in using Morans I for different time intervals to detect
disease clusters.
Ultimately I would like to use CUSUM - or similar monitoring statistic to
monitor the results of Morans I - similar to the work by
Rogerson (2005) Spatial Surveillance and Cummulative Sum Metho
On 17 November 2005 at 15:11, David Mitchell wrote:
| I didn't realise that the Windows version of R was "relocatable" that
| easily - I just assumed the install did something more complex than
| that. I'll try copying the R files to a USB key later today.
Yup. I've installed it onto Windows SMB
I use SLAX on my USB stick (1GB) and I carry R, a host of contributed
R packages, all the core BioC packages, rkward as the R frontend,
Octave, and many many more. SLAX is in my experience much smaller,
lighter, faster, and more complete than any of the other Linux Live
CDs out there (my i
Thanks guys,
Dirk: I generally resort to using my laptop for R analysis, but it
usually involves dragging loads of data multiple times between a
customer system and my laptop. Moving large amounts of data in this
fashion can be a problem, particularly when there's sensitivity issues
about the dat
Dear Sir/Madam,
I am a beginner in R. Here is my questions.
1. Can you give me one test for randomness (a name and descriptive
paragraph is sufficient).
2. I have learned a uniform random number generator [e.g. not the
algorithms: i)Wichmann-Hill, ii) Marsaglia-Multicarry, iii) Super-Duper
(Marsa
Dear R expert,
The following is my questions:
1. How to generate two sequences of 150 uniform
deviates, called v1 and v2, in the range [-1,1].
2. How to compute r=(v1)^2+(v2)^2
3.If r is outside the range of (0,1) then it will be
discarded.
4 How to compute (v1)*sqrt(-2*log(r)/r) and output th
On 17 November 2005 at 14:16, David Mitchell wrote:
| Hello list,
|
| A short time ago, I found
| http://johnhaller.com/jh/useful_stuff/portable_apps_suite/, which
| contains basically a complete set of office tools that can be run
| *entirely* from a USB key. The concept is:
| - find a Windows
David,
I've run old version of R (2 years ago) from a CD and it worked well for me.
I guess you could run R from a USB drive.
On 11/16/05, David Mitchell <[EMAIL PROTECTED]> wrote:
>
> Hello list,
>
> A short time ago, I found
> http://johnhaller.com/jh/useful_stuff/portable_apps_suite/, which
>
Hello list,
A short time ago, I found
http://johnhaller.com/jh/useful_stuff/portable_apps_suite/, which
contains basically a complete set of office tools that can be run
*entirely* from a USB key. The concept is:
- find a Windows PC
- put in your USB key
- run OpenOffice, Firefox, Gaim, Nvu, Thun
My data are (for one experiment):
3 Types of plant (1,2,3)
4 Species per Type (ex:for type 1(a,b,c,d), for type 2(e,f,g,h) and
for type 3(i,j,k,l)
8 Repetitions of each Species
3 Stages (10, 20, 30)
2 Measures per Stages, for the stages 20 and 30 (ex: for Stage 10(C),
Dear all,
Does anybody happen to have written some GEE source code in Fortran to
solve a Generalized Estimation Equation Model? Or kindly point out me a
good starting point. Thanks.
Zhu Wang
SCHARP
__
R-help@stat.math.ethz.ch mailing list
https://s
I managed to delete the original post about this and will no doubt mess up
threading of responses. The post is shown below.
This looks like a problem I saw just a couple of days ago. I noticed that
you used a path variable in determining the file location and name. If
something is wrong with y
This should work:
update.packages(ask = "graphics", repos = NULL,
contriburl = "file:///g:/myFolder/myRepository"))
-peter
Muhammad Subianto wrote:
> I try to update R packages via my local repository.
> I put all R packages in g:/myFolder/myRepository, I do like
>
>
>>library(tools)
>>wr
I try to update R packages via my local repository.
I put all R packages in g:/myFolder/myRepository, I do like
> library(tools)
> write_PACKAGES("g:/myFolder/myRepository")
> options(repos=c(LocalR="file://g:/myFolder/myRepository"))
> getOption("repos")
LocalR
"file://
>> Yes. The Statlib mirror (which I suppose is what you call PA2) appears to
>> be badly messed up.
>>
>> --
>>O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B
>> c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K
>> (*) \(*) -- University of Copenhagen Denm
Does this help?
http://cran.r-project.org/doc/FAQ/R-FAQ.html#How-do-I-produce-PNG-graphics-i
n-batch-mode_003f
Andy
From: Jack Zhu
>
> Hi all,
>
> When I ran a script containing the following codes:
>
> png(paste(savepath,"a_rnaplot.png",sep = ""),width =
> fwidth, height = fheight,points
Dear R-helpers,
I am working with a time series data set of 1000 observations and I am runnig
several models on it. I divided the sample in three sets: estimation [1:700],
validation, [701:990] and test [991:1000].
Now I was wondering if there is a more elegant way to do this.
Are
Hi all,
When I ran a script containing the following codes:
png(paste(savepath,"a_rnaplot.png",sep = ""),width = fwidth, height =
fheight,pointsize = fpointsize);
data_deg <- AffyRNAdeg(data_cel)
plotAffyRNAdeg(data_deg,col=cols,lty=1,lwd = "2")
#a <- par("fin")
legend("botto
"Tuszynski, Jaroslaw W." <[EMAIL PROTECTED]> writes:
> I do have single version of R (2.2.0) and all packages I have were installed
> by downloading precompiled packages from PA or NC repositories within last
> month.
>
> I also just managed to download 'modeltools' from PA2 and its DESCRIPTION
On Wed, 16 Nov 2005, tom wright wrote:
> I'm having a small problem using RODBC. I'm trying to retrieve a string
> from a very long memo field (512*20*9=9360 characters = 74880 bytes) in
> an MSAccess database.
> It appears that RODBC set a maximum buffer size for a single column of
> 65535 bytes.
On 16-Nov-05 Olivia Lau wrote:
> Hi,
>
> I'm trying to find a way to take evaluate the Normal CDF over an
> interval and return the result on the log scale. This works, but I
> think it isn't numerically stable:
>
> log(pnorm(a, mean = x, sd = y) - pnorm(b, mean = x, sd = y))
>
> Does anyone
I do have single version of R (2.2.0) and all packages I have were installed
by downloading precompiled packages from PA or NC repositories within last
month.
I also just managed to download 'modeltools' from PA2 and its DESCRIPTION
says:
Built: R 2.3.0; ; 2005-11-04 20:47:47; windows
R console:
On Wed, 16 Nov 2005, Liaw, Andy wrote:
> That's because S does partial matching of component names, if it can be
> done unambiguously:
>
>> z <- list(aaa=1, bbb=2)
>> z$a
> [1] 1
>> z[["a"]]
> [1] 1
>> z["a"]
> $"NA"
> NULL
[That could be done unambiguously, but [] does not partial match in R
(se
I'm having a small problem using RODBC. I'm trying to retrieve a string
from a very long memo field (512*20*9=9360 characters = 74880 bytes) in
an MSAccess database.
It appears that RODBC set a maximum buffer size for a single column of
65535 bytes.
## cut from RODBC.c ##
Hi,
I'm trying to find a way to take evaluate the Normal CDF over an interval and
return the result on the log scale. This works, but I think it isn't
numerically stable:
log(pnorm(a, mean = x, sd = y) - pnorm(b, mean = x, sd = y))
Does anyone know of a single function that does the above?
Tuszynski, Jaroslaw W. wrote:
> Uwe,
>
> I think I used following repositories: PA1, PA2 and NC, which are the
> closest to me and I think I got the same results in all. The package I
> downloaded was "coin" which downloaded other packages. The consol printout
> follows:
>
>
>
>>chooseCRANmirr
I am using the RODBC package to read data from an Excel file. An excerpt of
the file looks like this:
00103V206 AES Corporation 6.00% 42.87
00808N202 AES Trust III 6.75% 34.98
03748R861 Apartment Investment & Management 9.00% #ERROR (I)
039380209 Arch Coal, Inc.
Dear all,
I want to solve a score function by using Newton-Raphson algorithm. Is there
such a fucntion in R? I know there's one called optim, but it seems only doing
minimizing or maximizing.
Thanks,
Jimmy
__
R-help@stat.math.ethz.ch mailing lis
That's because S does partial matching of component names, if it can be done
unambiguously:
> z <- list(aaa=1, bbb=2)
> z$a
[1] 1
> z[["a"]]
[1] 1
> z["a"]
$"NA"
NULL
> z["aaa"]
$aaa
[1] 1
Andy
From: Hong Ooi
>
> Hi,
>
> I came across some rather unexpected behaviour the other day with
> a
Hello.
I have 16 subjects with 1-4 obs per subject.
I am using the package "nlme" to fit a simple random effects (variance
components model) with 3 parameters: overall mean (fixed effect),
between subject variance (random) and within subject variance (random).
I need a 3x3 variance-covariance
Thank you, that worked great. Per your suggestion, I transformed the
Precip data to fit on the same scale as the Temp data. After changing
trellis.par.set$layout.widths$right.padding to 5, I was able to fit a
second axis and label for the Precip (using panel.axis) to achieve the
plot I want.
E
Uwe,
I think I used following repositories: PA1, PA2 and NC, which are the
closest to me and I think I got the same results in all. The package I
downloaded was "coin" which downloaded other packages. The consol printout
follows:
> chooseCRANmirror()
> utils:::menuInstallPkgs()
trying URL
'http:
Tuszynski, Jaroslaw W. wrote:
> Hi,
>
> While installing precompiled packages I often get warnings like the one in
> the subject. I usually ignore them, but I still do not understand why
> windows packages are build with unreleased versions of R.
> Is there some way to get packages build under R
Jörg Schlingemann wrote:
> Hi!
>
>
>
> This is probably a very trivial question. Is there an easy way to invert the
> y-axis (low values on top) when using the function barplot()?
barplot(1:10, ylim=c(10, 0))
works for me ...
Uwe Ligges
>
>
> Thanks,
>
> Jörg
>
>
> [[alterna
I am using the package nlme to fit a simple random effects (variance
components model)
with 3 parameters: overall mean (fixed effect), between subject
variance (random) and
within subject variance (random).
I have 16 subjects with 1-4 obs per subject.
I need a 3x3 variance-covariance m
On Wed, 2005-11-16 at 16:46 +, Jörg Schlingemann wrote:
> Hi!
>
>
>
> This is probably a very trivial question. Is there an easy way to invert the
> y-axis (low values on top) when using the function barplot()?
>
>
>
> Thanks,
>
> Jrg
You mean something like this?:
barplot(1:10, y
Hi,
While installing precompiled packages I often get warnings like the one in
the subject. I usually ignore them, but I still do not understand why
windows packages are build with unreleased versions of R.
Is there some way to get packages build under R-2.2.0?
What are potential problems that ca
Perhaps use dput?
HTH,
Andy
From: Afshartous, David
>
> All,
>
> Usually when I save a variable I have ascii = FALSE since I usually
> load the variable via the load(variable) command and do not need to
> view the variable outside of R.
>
> When I set ascii = TRUE and view the variable outside
Em Qua 16 Nov 2005 11:30, jim holtman escreveu:
> What do you mean by 'abort'? Does an 'error' occur that you want to catch?
> If so, look at 'try'. Otherwise if it is testable, then just test for the
> condition and restart.
>
Hi,
this is not a real error, is a situation whitout a resolution.
I
Hi!
This is probably a very trivial question. Is there an easy way to invert the
y-axis (low values on top) when using the function barplot()?
Thanks,
Jörg
[[alternative HTML version deleted]]
__
R-help@stat.math.ethz.ch mailing list
You could try using functions like 'dput()' or 'dump()'. These may or may not
be equivalent to 'save()' depending on how complex the object to be saved is.
-roger
P.S. Asking people to respond to a different email address is not a good way to
get responses!
Afshartous, David wrote:
> All,
>
All,
Usually when I save a variable I have ascii = FALSE since I usually
load the variable via the load(variable) command and do not need to
view the variable outside of R.
When I set ascii = TRUE and view the variable outside of R in a text editor,
I notice additional characters (starting w/ RD
On Wed, 16 Nov 2005, Vivien W. Chen wrote:
> Hi!
>
> I am a beginner of R. I am trying to calculate integrate and draw a
> graph of the output, but just kept on getting error messages. I list my
> program and error message below. Please help. Many Thanks!
>
> ===
> + > a<--11
Hi there,
Is it possible to fix the x-axis on a dendrogram and allow vertical lines to
intersect to each other?
Regards,
Julio
-
[[alternative HTML version deleted]]
__
R-h
Effectively, it's much better
thanks
--- Prof Brian Ripley <[EMAIL PROTECTED]> a
écrit :
> On Wed, 16 Nov 2005, Florent Bresson wrote:
>
> > I have to compute some standard errors using the
> delta
> > method and so have to use the command
> "numericDeriv"
> > to get the desired gradient. Befor u
This is covered in the R News article.
On 11/16/05, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
>
>
> I was just looking for an easy way to convert between COM datetime and chron
> datetime (both ways.)
>
> I found examples on the list, but they involved origin.
>
>
>
> Does anyone have function
I was just looking for an easy way to convert between COM datetime and
chron datetime (both ways.)
I found examples on the list, but they involved origin.
Does anyone have functions for converting COM datetime <-> chron
datetimethat work "safely"?
David L. Reiner
> -Original Mes
(Ted Harding) <[EMAIL PROTECTED]> writes:
> On 16-Nov-05 Florent Bresson wrote:
> > I have to compute some standard errors using the delta
> > method and so have to use the command "numericDeriv"
> > to get the desired gradient. Befor using it on my
> > complicated function, I've done a try with a
You can also look at xyplot in the lattice package. You will have to set up
your data slightly differently than for the standard graphics package, but
it may well be worth learning to do so. The lattice package has enormous
flexibility for combining multiple sets of data in one panel or plotting
On 11/16/05 8:48 AM, "Vivien W. Chen" <[EMAIL PROTECTED]> wrote:
> Dear R users,
>
> If I have to combine plots which have the same independent and dependent
> variables in one graph. Which command should I use? Any example? Can I use
> "panel"?
Viven,
plot(x1,y1)
par(new=TRUE)
plot(x2,y2)
You
Dear R users,
If I have to combine plots which have the same independent and dependent
variables in one graph. Which command should I use? Any example? Can I use
"panel"?
Your help will be deeply appreciated. Thanks!
-Vivien Chen-
[[alternative HTML version deleted]]
Hi!
I am a beginner of R. I am trying to calculate integrate and draw a graph of
the output, but just kept on getting error messages. I list my program and
error message below. Please help. Many Thanks!
===
+ > a<--11
> b<-0.1
> c<-0.012
> x<-0:110
> t<-0:15
> integrand<-fu
Is this possible:
function(){
initialize <- function( ){ initialize }
for(i in test){
...
if( j <= test2 ) {i <- 1; initialize()}
...
}
Best,
Matthias
> Hi,
>
> I have the follow function:
>
> function() {
>
> ## Init of function
> ...
>
> for(i in test) {
> ...
>
On 16-Nov-05 Florent Bresson wrote:
> I have to compute some standard errors using the delta
> method and so have to use the command "numericDeriv"
> to get the desired gradient. Befor using it on my
> complicated function, I've done a try with a simple
> exemple :
>
> x <- 1:5
> numericDeriv(quot
Hi,
I have the follow function:
function() {
## Init of function
...
for(i in test) {
...
while(j <= test2) {
...
}
}
}
The problem is that sometimes, naturally, the while is not possible to be
resolved, and so the program abort.
In this case I need that program ret
On Wed, 16 Nov 2005, Florent Bresson wrote:
> I have to compute some standard errors using the delta
> method and so have to use the command "numericDeriv"
> to get the desired gradient. Befor using it on my
> complicated function, I've done a try with a simple
> exemple :
>
> x <- 1:5
> numericDe
Hello Martin,
2005/11/16, Martin Maechler <[EMAIL PROTECTED]>:
> Sorry to everyone -- I'm deeply embarrassed that this private
> reply (to Hans-Peter) accidentally went to the whole R-help list.
>
> It's bad enough (and inavoidable as long as we keep the list
> open) that real spam occasionally go
On 11/16/05, Robin Hankin <[EMAIL PROTECTED]> wrote:
> Hi
>
> I have a function f(k,l) which returns a matrix for integer k and l.
> I want to call f(.,.) for each combination of a supplied vector (as
> in expand.grid())
> and then I want to assemble these matrices into one big one using
> k and l
On Wed, 16 Nov 2005, Christopher Willmot wrote:
> The hist() command produces this message on my machine...
>
> Error in title(main = main, sub = sub, xlab = xlab, ylab = ylab, ...) :
> X11 font at size 14 could not be loaded
>
> How can I either (a) determine what font is required,
> or
I have to compute some standard errors using the delta
method and so have to use the command "numericDeriv"
to get the desired gradient. Befor using it on my
complicated function, I've done a try with a simple
exemple :
x <- 1:5
numericDeriv(quote(x^2),"x")
and i get :
[1] 1 8 27 64 125 21
a crude approach is the following:
f <- function(x){
k <- x[1]
l <- x[2]
matrix(k + l, k , l)
}
a <- c(1, 3)
b <- c(2, 4, 5)
combs <- expand.grid(a, b)
do.call("cbind", lapply(split(combs, combs$Var2), function(x)
do.call("rbind", apply(x, 1, f
I hope it helps.
Best,
Dimitris
[EMAIL PROTECTED] writes:
> On Wed, 16 Nov 2005, Peter Dalgaard wrote:
>
> > Torsten Hothorn <[EMAIL PROTECTED]> writes:
> > [snip]
> >
> > > > > However, how do I get Z from a Wilcoxon test in R?
> > > >
> > > > wtest <- wilcox.test(y~group,data=d, alternative="greater")
> > > > qnorm(wtest$p.va
On Wed, 9 Nov 2005, Arnaud Dowkiw wrote:
> Dera R helpers,
>
> I have a (x,y,z) data file where x and y are spatial coordinates and z a
> variable. I have some missing values in the z column and I would like to
> replace them with an optimized estimation from the neighbour cells. I could
> not
On Wed, 16 Nov 2005, Christopher Willmot wrote:
> The hist() command produces this message on my machine...
>
> Error in title(main = main, sub = sub, xlab = xlab, ylab = ylab, ...) :
> X11 font at size 14 could not be loaded
>
> How can I either (a) determine what font is required,
>
On Wed, 16 Nov 2005, Koen Hufkens wrote:
> Hi list,
>
> Is there a package that covers the evaluation of spatial statistics on
> images and not on point data? I've converted an image matrix to x, y
> coordinates and a measurement value but evaluation with the package
> spdep (not really design
Michael Graber wrote:
> Dear R- list,
> I have some data to present with histograms. Therefore I used hist(...).
> I have few values with almost 80% of
> the frequencies (totaly 800) and some other values with low frequencies
> ( totaly 5 -10 )
> that I want to emphasize. Therefore I want to "cut
Dear R-users,
I would need to do a Friedman test with replicated samples, and
Friedman.test(y...) currently works only for unreplicated designs.
Is there a script or a function available?
Thanks,
Alessandro
--
--
__
R-help@sta
Hi everyone,
before I embark in some coding, can I ask if there is any function in
any of the packages on CRAN dealing with the natural selection
coefficient S?
At the moment I am more curious to see it has been implemented already
in R, rather than what specific implementation.
Regards,
Federi
On Wed, 16 Nov 2005, Peter Dalgaard wrote:
> Torsten Hothorn <[EMAIL PROTECTED]> writes:
> [snip]
>
> > > > However, how do I get Z from a Wilcoxon test in R?
> > >
> > > wtest <- wilcox.test(y~group,data=d, alternative="greater")
> > > qnorm(wtest$p.value)
> > >
> >
> > or
> >
> > library("coin"
Sorry to everyone -- I'm deeply embarrassed that this private
reply (to Hans-Peter) accidentally went to the whole R-help list.
It's bad enough (and inavoidable as long as we keep the list
open) that real spam occasionally goes through, but
of all people, I should know better and be more careful
The hist() command produces this message on my machine...
Error in title(main = main, sub = sub, xlab = xlab, ylab = ylab, ...) :
X11 font at size 14 could not be loaded
How can I either (a) determine what font is required,
or (b) specify one of the fonts I have available?
This proble
Torsten Hothorn <[EMAIL PROTECTED]> writes:
[snip]
> > > However, how do I get Z from a Wilcoxon test in R?
> >
> > wtest <- wilcox.test(y~group,data=d, alternative="greater")
> > qnorm(wtest$p.value)
> >
>
> or
>
> library("coin")
> statistic(wilcox_test(y ~ group, data = d, ...), type = "stand
Hi
I have a function f(k,l) which returns a matrix for integer k and l.
I want to call f(.,.) for each combination of a supplied vector (as
in expand.grid())
and then I want to assemble these matrices into one big one using
k and l to index the position of the submatrices returned by f(k,l).
To
Hi list,
Is there a package that covers the evaluation of spatial statistics on
images and not on point data? I've converted an image matrix to x, y
coordinates and a measurement value but evaluation with the package
spdep (not really designed for image data I suppose) is unworkable. Any
sugge
> "Hans-Peter" == Hans-Peter <[EMAIL PROTECTED]>
> on Tue, 15 Nov 2005 16:10:35 +0100 writes:
Hans-Peter> Am 14.11.05 schrieb [EMAIL PROTECTED] <[EMAIL PROTECTED]>:
>> [snip some spam]
Hans-Peter> Being located only some blocks away from these spammers, I just
asked
On Tue, 15 Nov 2005, Thomas Lumley wrote:
> I think you have misinterpreted the direction of Peter's scepticism.
> The question is whether an effect size defined this way means anything
> useful. For example, even if your data are Normally distributed with
> equal variance this definition will not
On Tue, 15 Nov 2005, Thomas Lumley wrote:
> On Tue, 15 Nov 2005, Claus Atzenbeck wrote:
>
> > On Tue, 15 Nov 2005, Peter Dalgaard wrote:
> >
> However, for Wilcoxon tests, the formula for effect sizes is:
> r = Z / sqrt(N)
>
> I wonder how I can calculate the Z-score in R for
83 matches
Mail list logo
