Please use an informative subject line.
Your reference is vague. There have been many versions of locfit, and
quite a lot of code in earlier documentation no longer works, especially
that which precedes Loader's book.
library(locfit)
data(geyser)
fit.of -
Hi Harsh
As indicated in the answer to your first post, it is not so easy
to debug code without the code itself.
1) Have you checked the names of your variables so that you are
sure that they are correct. For example undesired white space
due to the automatic creation of variable names
On Mon, 13 Mar 2006, Martin Maechler wrote:
JohnF == John Fox [EMAIL PROTECTED]
on Sat, 11 Mar 2006 13:29:34 -0500 writes:
JohnF Dear Deepayan, As you say, it's currently very easy
JohnF to change settings (which is what I do routinely),
JohnF but since you asked, I much prefer
Hello Stephen,
As far as I know, the meta package will not allow you to include moderator
variables in the model. However, I have written a script for R/S-Plus that will
allow you to fit such models (essentially, these are mixed-effects models with
a random intercept). You can find the script
Thank you for the response.
Indeed, my subject line was ill-specified. My apologies.
Yes, it seems as if flim should be xlim; also, I should have just
added the same to the plot call.
Jacob
Prof Brian Ripley [EMAIL PROTECTED] 03/13/06 9:58 AM
Please use an informative subject line.
Your
Hello,
This query pertains to the application of funtion stl
I am new to R and working on a daily time series data which shows
periodicity every 12 months.I use the commands,
C-read.csv(C:/Charlotte.csv) # import data
D-as.Date(as.character(C[,1]),format=%m/%d/%Y)
You have not set up a time series! See ?ts.
You will in any case have a problem, as there are not a constant
number of days in a year.
On Mon, 13 Mar 2006, Amogh Deshpande wrote:
Hello,
This query pertains to the application of funtion stl
I am new to R and working on a daily time series
_
From: Vera, Graciela : Insurance Services (BISCO)
Sent: 10 March 2006 14:47
To: '[EMAIL PROTECTED]'
Subject: RE: .First functin
I have not used R for some time. Please accept my apologies if my question
has an obvious answer. I am trying to use R. For convenience and due to
Two questions about principal components analysis in R:
Q.1) Hogenraad and McKenzie (1999) used Bruce Thompson's FACSTRAP
program to bootstrap the factor loadings and scores in a principal
components analysis. The input to the analysis was a word-word
correlation matrix derived from a frequency
_
From: Vera, Graciela : Insurance Services (BISCO)
Sent: 10 March 2006 14:47
To: '[EMAIL PROTECTED]'
Subject: RE: .First functin
I have not used R for some time. Please accept my apologies if my question
has an obvious answer. I am trying to use R. For convenience and due to
When fitting a simple linear or polynomial regression using lm, R
provides a p-value for the whole model as well as for the individual
coefficients. When fitting the same models using gls (in order to
correct for autocorrelation), there doesn't seem to be a p-value
provided for the whole model,
Hello
I need to define complex likelihoods in BRUGS- mixed multinomial
probit random coefficient tobit for panel data
can I use the likelihood that i`ve written in R in the BRUGS model?
if not, how can I deal with complex likelihoods in BRUGS, there is no
integrate function in BRUGS?
is
Vera, Graciela : Insurance Services (BISCO) wrote:
_
From: Vera, Graciela : Insurance Services (BISCO)
Sent: 10 March 2006 14:47
To: '[EMAIL PROTECTED]'
Subject: RE: .First functin
I have not used R for some time. Please accept my apologies if my question
has
Johannes Reichl wrote:
Hello
I need to define complex likelihoods in BRUGS- mixed multinomial
probit random coefficient tobit for panel data
can I use the likelihood that i`ve written in R in the BRUGS model?
This is a (Win/Open)BUGS question which is not related to R (except that
you could use anova(fit1, fit2), where fit1 represent a model only
with intercept; however, you have to be a little bit more careful in
this setting since gls() has as default method the REML. Thus you
should specify, e.g.,
fit1 - gls(y ~ 1, correlation = ..., method = ML)
fit2 - gls(y ~ ...,
On Mon, 13 Mar 2006, Uwe Ligges wrote:
[...]
Vera, Graciela : Insurance Services (BISCO) wrote:
[...]
options(digits=5, length=999) # custom numbers and printout
What is option length doing? I do not know it.
It exists in S, and is the page length (in lines). In S, print() formats
output
That looks like it might do the job. I did not know about RSiteSearch(). Thanx
On 3/13/06, Charles C. Berry [EMAIL PROTECTED] wrote:
Simon Knapp sleepingwell at gmail.com writes:
Hi List,
I can't seem to find a package that implements boosting for
multi-class classification.
Does
Is there a function in R for constrained linear least squares?
I used the matlab function LSQLIN: my aim is to obtain
non-negative regression coefficients which sum 1.
Thanks in advance,
domenico vistocco
___
Yahoo! Mail:
A.J. Rossini wrote:
SNIP
One nice thing about ESS/Emacs is that it will cean a transcript
buffer. So if you are working in the inferior process buffer (where
you shouldn't be, but that's for another day), then you can write it
to disk as a transcript file, and clean it into a script.
Okay,
Domenico Vistocco vistocco at unicas.it writes:
Is there a function in R for constrained linear least squares?
I used the matlab function LSQLIN: my aim is to obtain
non-negative regression coefficients which sum 1.
Thanks in advance,
domenico vistocco
I haven't tried it, but it
Work in an inferior R session
M-x write-buffer to TonyIsSilly.Rt
So now your buffer is associated with a file, TonyIsSilly.Rt
M-x R-transcript-mode
M-x ess-transcript-clean-buffer
and you should be left with a script file, not a transcript file.
Enjoy. I'll be talking about such things
A.J. Rossini wrote:
Work in an inferior R session
M-x write-buffer to TonyIsSilly.Rt
So now your buffer is associated with a file, TonyIsSilly.Rt
M-x R-transcript-mode
M-x ess-transcript-clean-buffer
and you should be left with a script file, not a transcript file.
Enjoy.
I have some data which I would like to display with dotchart. The
labels are very long, so the chart becomes too small. Setting cex=0.7
seems to be a good compromise, but the gap between the text and the
chart still is too large. I did not find a gap parameter in the
description of dotchart...
On 3/13/06, Dietrich Trenkler [EMAIL PROTECTED] wrote:
I have some data which I would like to display with dotchart. The
labels are very long, so the chart becomes too small. Setting cex=0.7
seems to be a good compromise, but the gap between the text and the
chart still is too large. I did not
On 3/13/2006 9:33 AM, Dietrich Trenkler wrote:
I have some data which I would like to display with dotchart. The
labels are very long, so the chart becomes too small. Setting cex=0.7
seems to be a good compromise, but the gap between the text and the
chart still is too large. I did not find a
Hello.
I am an nxn data frame in the variable frame.
I want to make a contour plot with it. That is, I want to plot a square
of dimensions [1,n]x[1,n] with the gray level of square [i,i+1]x[j,j+1]
equal to frame[i,j].
How can I make it?
Thanks,
Arnau.
S-PLUS ESSENTIALS
24-27th April 2006
Venue : Insightful UK Office, Network House, Basing View, Basingstoke, Hants
RG21 4HG
We are pleased to be taking registrations for the next scheduled S-PLUS
Essentials course being held in Basingstoke 24-27th April 2006. If you are
unfamiliar with S-PLUS
Analysis of Microarray Data Using S-PLUS, R and S+ArrayAnalyzer
23rd March 2006
Presenter: Adam Diaz
Venue: Insightful UK, Network House, Basing View, Basingstoke, Hants RG21 4HG
This is the last chance to register for this unique course being run by a
member of Insightful's microarray
Hello,
I am currently testing a Vector AR of dim 3 over not a lot of data (135
* 3 observations) . To test the adequation of my vecot ar, I use the
Schwarz Bayesian Criterion and the classic modified Portmanteau test on
the residuals (it can be found for instance in
You could also consider using the quantreg package. The linear quantile
regression estimator rq() in this package has a weights option and using
this
linear model approach allows you to include covariates if necessary and
perform hypothesis tests and estimate confidence intervals. Nonlinear
Arnau Mir Torres wrote:
Hello.
I am an nxn data frame in the variable frame.
I want to make a contour plot with it. That is, I want to plot a square
of dimensions [1,n]x[1,n] with the gray level of square [i,i+1]x[j,j+1]
equal to frame[i,j].
How can I make it?
See ?image
Uwe
Hi!
I've just updated my version of R, from 2.0.1 to 2.2.1.
When I run my old lmer models in the new version, I get no p-values for
the fixed factors, i got them with the older version of R.
Also, when I try to run a binomial model in lmer, with the AGQ method I
get the following message:
Error in
Sorry that I send this text for a second time - I didn't see my posting
before (though I have set the preferences to see them)
If you want a comparison of different distance measures you might try
the manuscript of Johann Bacher who is a specialists in cluster methods.
He published an in-depth
Richard,
Since you mentioned that you can reformat the data any way you want,
I'll add a suggestion.
If you format your date/time data using R's default format, then your
R script can be a little simpler.
2006-03-11 15:09,0.014652
(i.e., put the date and time together in one column, using
Hi r-helpers,
When I issue the command
latex(anova(raw1.lmer0, raw1.lmer, raw1.lmerI), file = 'raw1.tex',
rownamesTexCmd = c('baR', 'addit', 'multip'), longtable = F, dcolumn
= T, booktabs = T, t able.env = F, colheads = NULL, colnamesTexCmd = c
('', 'df', 'aic', 'bic', 'logl', 'chisq',
Folks:
A question for R users on Macs. AFAIK, R's Tcl/TK tools allow one to build
GUI interfaces to R's functionality across all platforms. My question: How
well do the TK widget tools work on a Mac? More specifically, is there any
extra difficulty or complication in porting such applications to
Hi
I used R for the first time yesterday. I wanted to plot the aliasing
effect of sampling a 5.5KHz sinusoid at only 8KHz (below the Nyquist
limit). So I wrote a small R script that a) plots 1msec worth of a
5.5KHz sin wave b) plots 1msec of the resulting 2.5KHz alias and c)
plots the 8 sampling
Dear Erlend
For an answer to your first question about the degrees of
freedom, have a look at:
https://stat.ethz.ch/pipermail/r-help/2006-January/016384.html
To answer the second question: I think that the AGQ method is
not yet implemented as the error message implies. lmer is a
function that
Your error:
If you use plot(), the coordinate system of user coordinates is set up
each time, but you do want to plot in the coordinate system of your
first plot, hence use:
plot(time, signal, type = l, col = blue, xaxs = r, yaxs = r,
xlab = Time (msec), ylab = Signal, main = Aliasing,
That also works and is even more concise.Many thanks,
Paul
Uwe Ligges wrote:
Your error:
If you use plot(), the coordinate system of user coordinates is set up
each time, but you do want to plot in the coordinate system of your
first plot, hence use:
plot(time, signal, type = l, col =
-Original Message-
From: Deepayan Sarkar [mailto:[EMAIL PROTECTED]
Sent: Friday, March 10, 2006 4:10 PM
To: Andy Bunn
Cc: R-Help
Subject: Re: add trend line to each group of data in: xyplot(y1+y2 ~ x |
grp...
On 3/10/06, Andy Bunn [EMAIL PROTECTED] wrote:
Although this should
(This code looks vaguely familiar.)
Is anyone interested in participating in an effort to make a self
contained package for numerical derivatives? I would be happy to extract
the Richardson extrapolation code for first and second derivatives from
my package in the devel area of CRAN, but I'm a
Michael Kubovy kubovy at virginia.edu writes:
When I issue the command
latex(anova(raw1.lmer0, raw1.lmer, raw1.lmerI), file = 'raw1.tex',
rownamesTexCmd = c('baR', 'addit', 'multip'), longtable = F, dcolumn
= T, booktabs = T, t able.env = F, colheads = NULL, colnamesTexCmd = c
Dear R-users,
I'd like to announce the new version of package 'ltm' for Item
Response Theory analysis. The function grm() (along with supporting
methods, i.e., anova, margins, factor.scores, etc.) has been added for
fitting the Graded Response Model for ordinal polytomous manifest
variables.
On 3/13/06, Andy Bunn [EMAIL PROTECTED] wrote:
Deepayan:
I'm really trying to understand how to access the different groups in this
type of plot. Imagine I have the same scenario as above, an xyplot of the
form y1+y2 ~ x | grp. Now, how can I modify the call to panel.text to write,
say, the
Hi All,
I'm looking for some hints on idiomatic R usage using 'lapply' or similar.
What follows is a simple example from which to generalize my question...
# Suppose, in this simple example, I want to plot a number of different
lines in different colors;
# I define the
Hello. I'm having difficulties getting R_alloc() to work.
The test platform is a 64-bit Mac running R 2.2.1 installed via
a .dmg file obtained from the R site. Details are given below, in a
level of detail that I hope is appropriate. My eye was particularly
drawn to line #2 in the gdb
Try this:
plot(Y ~ X, d, type = n)
f - function(i) abline(lm(Y ~ X, d, subset = D == i), col = colors[i+1])
junk - lapply(unique(d$D), f)
On 3/13/06, John McHenry [EMAIL PROTECTED] wrote:
Hi All,
I'm looking for some hints on idiomatic R usage using 'lapply' or similar.
What
Hi all R-helpers!!!
Im a Newbie to R and I have only one small question:
Is there any possibility to plot a cmeans (package
e1071) cluster result? Or any other possibility of
visualisation like with the algorithm fanny? I
couldnt find any information how to create a plot
with cmeans on the R
Hi Paul, Tolga, and others:
I had also written some codes to compute derivatives, jacobians, and
Hessians. Please see the attached file for the code. I will be happy to
help out with the development of a package and/or with the documentation
process.
Best,
Ravi.
-Original Message-
Note that there is a Task View for ecology at:
http://cran.miscellaneousmirror.org/src/contrib/Views/Environmetrics.html
On 3/13/06, Dave Roberts [EMAIL PROTECTED] wrote:
In addition to the references from Professor Ripley, you might be
interested in the R packages and pages maintained by
On Mon, 13 Mar 2006, John McHenry wrote:
Hi All,
I'm looking for some hints on idiomatic R usage using 'lapply' or similar.
What follows is a simple example from which to generalize my question...
# Suppose, in this simple example, I want to plot a number of different
lines in
Hi all,
Why
fr2-function(x)
{ if(x1){x*x}else{1}
}
integrate(fr2,-1,2)
gives the following wrong answer:
3 with absolute error 3.3e-14
Warning message:
the condition has length 1 and only the first element will be used in: if (x
1) {
while
fr-function(x){as.numeric(x1)*(x^2-1)+1}
Le 13.03.2006 15:58, Arnau Mir Torres a écrit :
Hello.
I am an nxn data frame in the variable frame.
I want to make a contour plot with it. That is, I want to plot a square
of dimensions [1,n]x[1,n] with the gray level of square [i,i+1]x[j,j+1]
equal to frame[i,j].
How can I make it?
The function need to be vectorized. Try:
fr2.new - function(x) ifelse(x 1, x * x, 1)
integrate(fr2.new, -1, 2)
1.67 with absolute error 3.4e-05
Andy
From: Lynette Sun
Hi all,
Why
fr2-function(x)
{ if(x1){x*x}else{1}
}
integrate(fr2,-1,2)
gives the following wrong
In addition to the references from Professor Ripley, you might be
interested in the R packages and pages maintained by ecologists for such
work (even if you're doing movies). Packages labdsv, vegan, and ade4
both have a broad variety of distance/dissimilarity indices and numerous
alternative
When I try to run the example from Variogram with an lme object, I get
an error (although summary works):
R : Copyright 2005, The R Foundation for Statistical Computing
Version 2.2.1 (2005-12-20 r36812)
ISBN 3-900051-07-0
...
fm1 - lme(weight ~ Time * Diet, BodyWeight, ~ Time | Rat)
Error:
Thanks Dieter,
Unfortunately this doesn't work.
rowname = NULL
suppresses row names altogether. I want to substitute different row
names by writing
rownamesTexCmd = c('baR', 'addit', 'multip')
When I do this without writing rowname = NULL, my row names appear
first, followed by the
Hi there,
For the hierarchical clustering methods in R, are there any of them are
applicable for the large dataset? Coz one of the main shortcomings of
the hierarchical clustering methods is not good for large dataset. The
commands I found are hclust, agnes, and Diana. Is any of them good
for
Linda Lei wrote:
Hi there,
For the hierarchical clustering methods in R, are there any of them are
applicable for the large dataset? Coz one of the main shortcomings of
the hierarchical clustering methods is not good for large dataset. The
commands I found are hclust, agnes, and
Hi,
I'm trying to do a surface plot using the wireframe function.
Everything is working beautifully except that I want to be able to
re-scale it a LOT so that I can fit ~145 labels on the x-axis or y-axis.
I've tried using zoom, scales, aspect, .. nothing seems to work. The
help on wireframe in
Hi,
the function 'fr' evaluates each element of the vector
'x', independently of its length. 'fr2' does not.
Try with
fr2-function(x){
ifelse(x1, x*x, 1)
}
integrate(fr2,-1,2)
1.67 with absolute error 3.4e-05
Marco
--- Lynette Sun [EMAIL PROTECTED] wrote:
Hi all,
Why
Hi, Paul, Tolga, Ravi, et al.:
What about using splines to compute both derivatives and integrals?
These could be used, for example, to summarize complicated, expensive
Monte Carlo results in a relatively simple form could then be
manipulated analytically and quickly converted to
Hello,
I am trying to do model autocalibration using the snow and rgenoud
packages. The function I want to run in task-parallel fashion across
multiple machines is one that pre- and post-processes data and runs an
external model code. My problem is that external file I/O is happening
only in
Je serai absent(e) du 13/03/2006 au 20/03/2006.
Je répondrai à votre message dès mon retour.
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
Hi Scott --
It took me a bit to figure it out, but the help page for system made
it seem like system should return the exit status, rather than the
command result, if system is invoked without specifying intern =
TRUE. So why does system(hostname) actually print the host name?
it's a side-effect,
dear R wizards---I must be standing on the line (German
translation---this must be obvious). This is probably embarrassingly
obvious to anyone, except to me right now. maybe it is late.
I would like to run my R code in batch, but not see the R compilations
and my R source code---I just want to
Hello,
Having searched the archives and Google, I'm reasonably confident
that this is a new question, though I admit to being a new R user and thus
potentially missing something simple.
I have been using the sem package, and am interested in creating a
mean structure model (e.g.,
On Mon, 2006-03-13 at 20:17 -0500, ivo welch wrote:
dear R wizards---I must be standing on the line (German
translation---this must be obvious). This is probably embarrassingly
obvious to anyone, except to me right now. maybe it is late.
I would like to run my R code in batch, but not see
Hi ,
I am using nls(formula)
where formula is y~ expression
y is 1 character
~ is 1 character
now i have an expression that is greater tha 502 characters. but R does not
recognize anything greater than 502 characters. (Note the expression itself is
created using paste
Hi ,
I am using nls(formula)
where formula is y~ expression
y is 1 character
~ is 1 character
now i have an expression that is greater tha 502 characters. but R does not
recognize anything greater than 502 characters. (Note the expression itself is
created using paste
hi marc: yes, this is it. I tried --slave alone, which had not been
enough. thanks. regards, /iaw
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
I sent this to Bioconductor, but I think it is more of an R listserv
question, because of memory allocation trouble
I am having trouble getting the function heatmap() to work on the following
gene expression
dim(SAMPLES_log)
[1] 1262620
sample1
heatmap() will do hierarchical clustering with hclust(), and to do that it
needs to compute and store the lower triangular part of the distance matrix,
which has 12626 * 12625 / 2 elements, each elements taking up 8 bytes of
storage. That works out to about 608MB (and is about the size that R
Hi ,
I am using nls(formula)
where formula is y~ _expression
y is 1 character
~ is 1 character
now i have an _expression that is greater tha 502 characters. but R does not
recognize anything greater than 502 characters. (Note the _expression itself
is created using
On Mon, 2006-03-13 at 07:50 +, Prof Brian Ripley wrote:
`Ordination' is ecologists' terminology for multidimensional scaling.
You will find worked examples in MASS (the book, see the R FAQ), and the
two most commonly used functions, isoMDS and sammon, in MASS the package.
'Ordination' in
So this is most likely a bug in package nlme. However, we need a
reproducible example to be able to do anything about it, and without even
the traceback() we cannot be sure that it is in nlme.
Please follow the bug-reporting procedure.
On Mon, 13 Mar 2006, Rick Bilonick wrote:
When I try
Michael Kubovy kubovy at virginia.edu writes:
Unfortunately this doesn't work.
rowname = NULL
suppresses row names altogether. I want to substitute different row
names
This example works for me. Take into account that rownamesTexCmd adds
the leading \, so you cannot use it
Rick Bilonick rab45+ at pitt.edu writes:
When I try to run the example from Variogram with an lme object, I get
an error (although summary works):
library(nlme)
fm1 - lme(weight ~ Time * Diet, BodyWeight, ~ Time | Rat)
Variogram(fm1, form = ~ Time | Rat, nint = 10, robust =
79 matches
Mail list logo