Hi,
When I place a legend on a plot it looks exactly as I intended
on the screen. However, almost always, when I export this to
postscript file, the legend's text protrudes through the legend's
frame (the latter being placed correctly).
See the appended example code. I can send the EPS file as
On Wed, Jan 14, 2004 at 03:18:10PM -0500, Liaw, Andy wrote:
If pam produces the cluster medoids, you should be able to use the
1-nearest-neighbor classifier for prediction of future data, using the
medoids as the `training' data. 1-NN is available in the `class' package,
part of the `VR'
Hi!
On Wed, Jan 14, 2004 at 11:08:44PM -0800, Itay Furman wrote:
When I place a legend on a plot it looks exactly as I intended
on the screen. However, almost always, when I export this to
postscript file, the legend's text protrudes through the legend's
frame (the latter being
Samuel == Samuel Kemp (Comp) [EMAIL PROTECTED]
on Wed, 14 Jan 2004 13:53:10 + writes:
Samuel Hi, I am trying to use the stl function in the ts
Samuel package. It requires that the data is a univariant
Samuel time series at the moment my data is in a vector. I
Samuel have
On Wed, 14 Jan 2004, Spencer Graves wrote:
Yes, but glm maximizes the binomial likelihood assuming
log(p/(1-p)) is a linear model. Therefore, you don't have to transform
the 0's and 1's. There are cases where a particular combination of
potential explanatory variables will clearly
The short answer is not to copy the device, but to replot on the new
device. That is the advice given in MASS, for example.
When you copy a device, you replay the device list and hence the lines and
text are placed at the positions calculated using the font metrics of the
first device and not
Hi
#-
# funkce pro automaticke oznaceni piku ve spektru (peaks)
# autor Brian Ripley
# span has to be odd number
peaks-function(series,span=3)
{
z - embed(series, span)
s - span%/%2
v- max.col(z) == 1 + s
result -
On Thu, Jan 15, 2004 at 08:32:45AM +, Prof Brian Ripley wrote:
On Thu, 15 Jan 2004, Renald Buter wrote:
On Wed, Jan 14, 2004 at 03:18:10PM -0500, Liaw, Andy wrote:
If pam produces the cluster medoids, you should be able to use the
1-nearest-neighbor classifier for prediction of
Benjamin == Benjamin STABLER [EMAIL PROTECTED]
on Wed, 14 Jan 2004 15:52:46 -0800 writes:
Benjamin So it looks like R stores numbers as doubles
Benjamin unless the are converted to integers (long) with
Benjamin the as.integer() function or they are created
On Wed, 14 Jan 2004 [EMAIL PROTECTED] wrote:
I am trying to figure out R data types and/or storage mode. For example:
[...]
So it looks like R stores numbers as doubles unless the are converted to
integers (long) with the as.integer() function or they are created with the
: operator. If
On Thu, 15 Jan 2004, Renald Buter wrote:
On Thu, Jan 15, 2004 at 08:32:45AM +, Prof Brian Ripley wrote:
On Thu, 15 Jan 2004, Renald Buter wrote:
On Wed, Jan 14, 2004 at 03:18:10PM -0500, Liaw, Andy wrote:
If pam produces the cluster medoids, you should be able to use the
I'm analizing the Argentina stock market (merv)
I download the data from yahoo
library(tseries)
Argentina - get.hist.quote(instrument=^MERV,1996-10-08,2003-11-03,
quote=Close)
merv - na.remove(log(Argentina))
I made the Augmented Dickey-Fuller test to analyse
if merv have unit root:
Dear All,
A friend asked me for help with a regression problem dealing with y and x, both
random variables.
The x and y are respectively the observed and estimated biomass values of a
green algae derived from an ecological model.
We first considered it to be a model II regression and
On Thu, Jan 15, 2004 at 08:59:37AM +, Prof Brian Ripley wrote:
[snip]
# separate the ruspini data into train and test set
train-ruspini[1:50,]
test-ruspini[51:75,]
pamx-pam(train,4)
knnx-knn(pamx$medoids,test,factor(c(a,b,c,d)),k=3)
knnx
[1] d d b b d c
Ok I made Jarque-Bera test to the residuals (merv.reg$residual)
library(tseries)
jarque.bera.test(merv.reg$residual)
X-squared = 1772.369, df = 2, p-value = 2.2e-16
And I reject the null hypotesis (H0: merv.reg$residual are normally
distributed)
So I know that:
1 - merv.reg$residual aren't
Hello,
by checking the precision of a convolution algorithm, we found the
following inexactness:
We work with R Version 1.8.1 (2003-11-21) on Windows systems (NT, 2000,
XP).
Try the code:
## Kolmogorov distance between two methods to
## determine P(Poisson(lambda)=x)
Kolm.dist - function(lam,
I am running R-1.81, apple's JVM installed with OS 10.3.2, and installed
Sjava-0.65.X (a patched version of Sjava for MacOS from Simon Urbanek, which
I know is inherently dangerous, but...). This version installed fine and
some aspects seem to work as expected. However, this is one of the
If I have 100 objekts in a folder and I prefer not to load them manually I
wonder how I do this in R if using a for-loop.
I was thinking initially to do something like this:
infiles - dir(pattern=.RData)
for(i in length(infiles))
{
load(infiles[i])
paste(kalle, i, sep=)
Johan Lindberg wrote:
If I have 100 objekts in a folder and I prefer not to load them manually
I wonder how I do this in R if using a for-loop.
I was thinking initially to do something like this:
infiles - dir(pattern=.RData)
For sure you mean
infiles - dir(pattern = \\.RData)
for(i in
Hi
I compouted a multiple linear regression with repeated measures on one
explanatory variable:
BOLD peak (blood oxygenation) as dependent variable,
and as independent variables I have:
-age.group (binaray:young(0)/old(1))
-and task-difficulty measured by means of the reaction-time 'rt'. For
You have two problems.
1) You want to use
assign(paste(kalle, i, sep=), saveLoadReference)
to create the name.
That one is discussed frequently enough to be an FAQ, and it is Q7.23.
2) You need to keep the return value of load to assign to the object.
So I think you want
for(i in
Well, I don't think this is ANCOVA as you seem to want to specify a random
slope for a covariate. aov() is not designed for that. It is also not
designed for assessing the size of fixed effects which seems the question
here.
As I understand it, you have only one observation for each value of
Renald == Renald Buter [EMAIL PROTECTED]
on Thu, 15 Jan 2004 12:22:18 +0100 writes:
Renald Thank you *very* much for your help. I thought I'd let the list know
Renald what I did to get it right:
# create a seed vector
seed-rank(runif(75))
S has a function for
Hello again. This is a resend (first sent on the 12th of January);
I would truly appreciate a response.
I have tried, unsuccessfully, to install the Hmisc and Design libraries
(Harrell FE Jr) in R-aqua (Apple). These libraries are in source code only
(i.e. not yet in ready-to-use binary
On Thu, 15 Jan 2004, Bema Bonsu wrote:
Hello again. This is a resend (first sent on the 12th of January);
I would truly appreciate a response.
I have tried, unsuccessfully, to install the Hmisc and Design libraries
(Harrell FE Jr) in R-aqua (Apple). These libraries are in source code only
by() will sort on the levels of `class', so I think what you need to do is
ensure that `class' has its levels sorted numerically. Try
df$class - factor(as.numeric(as.character(df$class)))
to do that. Then to sort the df by numerical class you could use
df - df[sort.list(unclass(df$class)), ]
I have a dataframe such that when I enter the dataframe name at the R prompt
and see the data, the order of the data is correct (ie. what I want -
ordered numerically by the factor, class). The table is akin to the
following:
df: (dataframe = df)
uniqueID class age
a 132
b
Hi!
How to generate a empty string of a precise length without using a loop?
I need a empty string (char **) to pass it with .C function to a c dll. The c
routine writes the result into the string.
Eryk.
Dipl. bio-chem. Eryk Witold Wolski@MPI-MG Dep. Vertebrate Genomics
On Thu, 15 Jan 2004, Wolski wrote:
Hi!
How to generate a empty string of a precise length without using a loop?
paste(rep( ,n), collapse=)
gives a string of n spaces.
-thomas
I need a empty string (char **) to pass it with .C function to a c dll.
The c routine writes the
Dear Christian:
That is not the problem, but thanks for your attempted help. I still
don't know what the problem is. Has anyone encountered this while
using gnls function in the package nlme?
thanks again for any help,
Ravi.
- Original Message -
From: Christian Mora [EMAIL
Thomas Lumley wrote:
paste(rep( ,n), collapse=)
gives a string of n spaces.
Is that more or less efficient/faster or slower than using sprintf:
sprintf(paste(%,n,s,sep=''), )
A quick test shows it to take about two-thirds the CPU time of the
paste(rep()) solution.
Not that it matters - I
Hi:
I tried Prof. Ripley's suggestion and I got the following error message:
ans52g - gnls(log(b52) ~ p0 + p1/(1 + exp(-(log(dose)-p2)/p3))^p4,
start=list(p0=3,p1=8,p2=2,p3=2,p4=1.5),control=gnlsControl(tol=1.e-07),
weights=varPower(form=~fitted(.)))
Error in varPower(form = ~fitted(.)) : form
I am looking for old version of R for Windows (before 1.5). Where would I
find them?
Michael
LEGAL NOTICE\ Unless expressly stated otherwise, this messag...{{dropped}}
__
[EMAIL PROTECTED] mailing list
Can you help me to overlap a histogram and theoretical density curve of
poison distribution?
There is more than one Poisson distribution you could possibly overlap but a
common choice is the Poisson with parameter = observed mean of the data - in
your case, 0.7.
I think the below is somewhat
Dear R experts:
Can you help me to overlap a histogram and theoretical density curve of poison
distribution? for example data like this:
The numbers of sanils found in each of 100 sampling quadrats in an area were
as follows:
number of snails, r 0 1 2 3 4 5 8 15
f=frequency of r 69
For a long time I've wanted a way to conveniently extract multiple elements
from a list, which [[ doesn't allow. Can anyone suggest an efficient
function to do this? Wouldn't it be a sensible addition to R?
For example,
alist - list()
alist[[1]] - list()
alist[[1]]$name - first
alist[[1]]$vec
The sources are on CRAN still, and older compilers (if needed) are
available via www.mingw.org. So where is the difficulty?
On Thu, 15 Jan 2004, Man, Michael wrote:
I am looking for old version of R for Windows (before 1.5). Where would I
find them?
--
Brian D. Ripley,
Aren't sapply()/lapply() sufficient for this?
sapply(alist, function(x) x$vec)
[,1] [,2]
[1,]15
[2,]26
[3,]37
[4,]48
lapply(alist, function(x) x$vec)
[[1]]
[1] 1 2 3 4
[[2]]
[1] 5 6 7 8
HTH,
Andy
From: Waichler, Scott R
For a long time I've wanted a
Hi,
There's probably some simple way of doing this, but I'm just not seeing
it - How do I get the background to be white instead of grey when I have
a cloud plot (using the lattices package)? par(bg=white) isn't
working. I'm assuming par commands won't work on lattice plots. What
should I use
Would it be smart to add a hyperlink for the R logo (in the help files) that
would go the, or a, R homepage?
---
Jeff D. Hamann
Forest Informatics, Inc.
PO Box 1421
Corvallis, Oregon USA 97339-1421
(office) 541-754-1428
(cell) 541-740-5988
[EMAIL PROTECTED]
www.forestinformatics.com
[] works on a list and extracts multiple elements. However, that is not
it seems what you want, rather, unions of elements of elements of lists.
That is a pretty unusual need, and perhaps you could explain you you need
it.
Andy's solution still need something like
do.call(c, lapply(alist,
I've been trying to plot the predicted values, as a line, over the data for
a simple nonlinear fit with the following commands:
plot( hg ~ ht )
... define some function hg ~ ht + junk ...
... blah, blah, obtain parameter estimates and predicted values, blah...
... then...
lines( sort( $predicted
For the first case, I actually do this pretty often and usually use
something like:
as.vector(sapply(alist, [[, vec))
For the second case, I think you just need to use lapply():
alist - lapply(seq(along = alist), function(i)
alist[[i]]$name - new.names[i])
-roger
Waichler,
Sorry, that second example should be
alist - lapply(seq(along = alist), function(i) {
alist[[i]]$name - new.names[i])
alist
})
-roger
Roger D. Peng wrote:
For the first case, I actually do this pretty often and usually use
something like:
From: Jeff D. Hamann
I've been trying to plot the predicted values, as a line,
over the data for
a simple nonlinear fit with the following commands:
plot( hg ~ ht )
... define some function hg ~ ht + junk ...
... blah, blah, obtain parameter estimates and predicted
values, blah...
Ich habe in R eine Version von SIR gefunden und ausprobiert. Leider kann
diese multivariate Responses nicht verarbeiten. Gibt es in R eine
ausgefeilte Version von SIR?
__
[EMAIL PROTECTED] mailing list
Ich habe in R eine Version von SIR gefunden und ausprobiert. Leider kann
diese multivariate Responses nicht verarbeiten. Gibt es in R eine
ausgefeilte Version von SIR?
Danke für die Hilfe.
V.H.
__
[EMAIL PROTECTED] mailing list
Brian described well the operation I would like to do.
I'm not familiar with do.call() but I'll work on that.
Yes, ideally I would like to access values throughout a list object
with fully implict indexing, such as the invalid alist[[1:2]]$vec[c(2, 4)].
Notice I was hoping to subset anywhere in
Is there a fonction for multiple comparison tests (similar to multicomp in Splus) in
a package of R?
Thanks in advance for any hint...
Cheers,
Patrick Giraudoux
University of Franche-Comté
Department of Environmental Biology
EA3184 af. INRA
F-25030 Besançon Cedex
tel.: +33 381 665 745
There is now a package in R called multcomp for general multiple
comparisons that does things similar to the Splus library you mentioned.
BTW, a search of the help archives for multicomp or multiple
comparisons brings this up.
HTH, Andy
__
[EMAIL
Dear Jeff,
I'm not sure that I follow entirely what you've done, but perhaps the
following suggestions will help: (1) If the plotted curve isn't smooth
because it's evaluated at too few x-values or at x-values that are too
unevenly spaced, what about getting a sufficient number of predicted
Dear Adrienne,
I'm aware of a couple of ways to get a white background in trellis
graphics. One is lset(col.whitebg()).
I hope that this helps,
John
At 07:04 PM 1/15/2004 +, Mueller, Adrienne wrote:
Hi,
There's probably some simple way of doing this, but I'm just not seeing
it - How do I
Try
trellis.par.set(background, list(col = 0))
Or you can explicitly launch the trellis device and set `bg = 0'.
-roger
Mueller, Adrienne wrote:
Hi,
There's probably some simple way of doing this, but I'm just not seeing
it - How do I get the background to be white instead of grey when I have
Hello,
I have written perl programs that extract data from a
text file, process them, and create other text files,
which I'd like to apply some statistics too (for
example with R).
I'd like to do it all in once , with a single script.
I'm not familiar with R, I'd like to know if this task
could
Hi,
How can I assign memory in R? My simulations are very slow.
Thanks!,
-R
[[alternative HTML version deleted]]
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
Hi all,
In the (very simple) example below, I have defined a random effect for
the residuals in lme(). So the model is equivalent to a FIXED effect
model. Could someone explain to me why lme() still gives two standard
deviation estimates? I would expect lme() to return either:
a) an error or a
Hi there -
Is there a package that computes an adjusted rand index?
Thanks a lot - Ton
Ton van Daelen, PhD
Director, Application Support
Tel: (858) 279-8800 ext 217
Fax: (858) 279-8804
Web: www.scitegic.com
Register now for the 2004 Pipeline Pilot user group meeting Jan 28-30 in San Diego:
Jerome Asselin [EMAIL PROTECTED] writes:
Hi all,
In the (very simple) example below, I have defined a random effect for
the residuals in lme(). So the model is equivalent to a FIXED effect
model. Could someone explain to me why lme() still gives two standard
deviation estimates? I would
I just installed Fedora in VMWare.
Can somebopdy tell me what lines i have to put in
yum.conf
so R will be automatically integrated in the package system
and updated when a new release is available?
--
Erich Neuwirth, Computer Supported Didactics Working Group
Visit our SunSITE at
On Thu, 2004-01-15 at 16:30, Douglas Bates wrote:
...snip...
(BTW, I wouldn't say that this is equivalent to a fixed effects
model. It is still a random effects model with two variance
components. It just doesn't have well-defined estimates for those two
variance components.)
Agreed.
-Original Message-
From: Enrico Curiotto [mailto:[EMAIL PROTECTED]
Hello,
I have written perl programs that extract data from a
text file, process them, and create other text files,
which I'd like to apply some statistics too (for
example with R).
I'd like to do it all in once
Waichler, Scott R wrote:
Brian described well the operation I would like to do.
I'm not familiar with do.call() but I'll work on that.
Yes, ideally I would like to access values throughout a list object
with fully implict indexing, such as the invalid alist[[1:2]]$vec[c(2, 4)].
Notice I
On Thu, 15 Jan 2004, Enrico Curiotto wrote:
Hello,
I have written perl programs that extract data from a
text file, process them, and create other text files,
which I'd like to apply some statistics too (for
example with R).
I'd like to do it all in once , with a single script.
I'm not
When I first started using lattice I found the colour schemes a bit
confusing. So eventually I came up with the colours I wanted.
The code below was one of those attempts. One thing that happened
however was that I kept shutting down the graphics window that pops up
and the colours would revert
64 matches
Mail list logo
