On Thu, 18 Mar 2004, Adrian Custer wrote:
In the course of trying to implement the CDF of an
InverseGammaDistribution, I have run across the need for an igamma()
function. Several others have needed this function but the answers I
have found so far are not totally clear to me. I'm writing for
[EMAIL PROTECTED] writes:
Why is there an error on profile, which should only evaluate the
objective function for different parameter values?
profile() doesn't merely evaluate the objective function at
different parameter values. It holds one of the parameters constant
and optimises the
I've been reading the R mail archives and I've found a lot of messages
with this same kind of problem, but I can't understand the answers. Can
one of you try to explain this to me?
Here's my example. Given a regression model and a variable, I want to
use unstack() on the vector of residuals
Teresa Alarcon wrote:
Hi!
Is there exists an instruction to read images (bmp, jpg, etc.) in R?
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
On Thu, 18 Mar 2004, Randy Zelick wrote:
Is there a way to *not* supress leading zeros when printing (to the
console window or to a file) a dataframe?
If you mean via print() or autoprinting, no.
I am not sure why you would want to do this, but it seems that using
format() and then gsub should
Adrian == Adrian Custer [EMAIL PROTECTED]
on Thu, 18 Mar 2004 18:47:54 -0800 writes:
Adrian Hello all,
Adrian In the course of trying to implement the CDF of an
Adrian InverseGammaDistribution, I have run across the need for an igamma()
Adrian function. Several others have
Dear users
I would like to detect and estimate common patterns in a set of time
series. I have heard about the Dynamic Factor Analysis (DFA) which is a
multivariate time series analysis method to perform such analysis.
Does anybody know if there is a librairy dedicated to DFA?
Thank in
Note corrections to two typos, since Master Custer seems to get confused
by tiny details.
On Fri, 19 Mar 2004, Prof Brian Ripley wrote:
On Thu, 18 Mar 2004, Adrian Custer wrote:
In the course of trying to implement the CDF of an
InverseGammaDistribution, I have run across the need for an
Randy == Randy Zelick [EMAIL PROTECTED]
on Thu, 18 Mar 2004 21:00:08 -0800 (PST) writes:
Randy Hello list,
Randy Is there a way to *not* supress leading zeros when printing (to the
Randy console window or to a file) a dataframe?
Yes, e.g. use
formatC(..., format = f)
to
Paul Johnson [EMAIL PROTECTED] writes:
I've been reading the R mail archives and I've found a lot of
messages with this same kind of problem, but I can't understand the
answers. Can one of you try to explain this to me?
Here's my example. Given a regression model and a variable, I want to
This is exactly what I mean by PPDE. I was reading the paper you mention below
and I was looking for already implemented code in languages like R, Matlab or
C.
Thank you all
Luis
I don't think anyone has actually answered this. PPDE is a technical
term, defined in
Friedman, Steutzle
The problem is the use of a formula, and hence non-standard evaluation
(even non-standard for a formula). Unstack contains
res - c(tapply(eval(form[[2]], x), eval(form[[3]], x), as.vector))
and those eval() calls are looking only in x and then along the search
path. They should be
On 19 Mar 2004, Peter Dalgaard wrote:
Paul Johnson [EMAIL PROTECTED] writes:
I've been reading the R mail archives and I've found a lot of
messages with this same kind of problem, but I can't understand the
answers. Can one of you try to explain this to me?
Here's my example.
In function daystoyears in package pastecs, I get this (wrong?) result
with 1995:
daystoyears(1,datemin=1/1/1997,dateformat=m/d/Y)
1997.001
daystoyears(1,datemin=1/1/1995,dateformat=m/d/Y)
1994.999
Any insights?
Thanks
Angel
__
I have been asked how to handle the following situation in R:
Given an unbalanced design of 3 crossed random effects, such as
subject, rater and item, how to estimate the variance components?
I know how to do it using lme, but this seems to be limited to
the nested case; or to use aov with error
Hi
Is there a way to draw a line across several graphs?
Say I have a few histograms one upon the other with 0 exactly on one line.
Now when I do abline(v=0, lty=2) I have only the short lines
corresponding to each graph.
How can I draw one line through?
Thanks,
David
Dear R-ers,
I am just a simple 'end-user' of R and am trying to analyse data with a binary
response variable (dead or alive) in relation to weight and sex (of young birds). As
some of the birds have the same biological mother, I am using mixed models with the
identity of the mother as a random
This has probably been addressed here, but I can't find it if it has,
and it may be up on the R project homepage or the CRAN pages, but if
so
... so sorry if this has been answered before: question is, can I
upgrade to 1.8.1 under Debian stable (a.k.a. Woody)? The deb
packages in the
I prefer to use R on a linux box and ultimately need things to end up
there to serve things up using David Firth's excellent CGIwithR and
apache, but one step at a time and I've installed 1.8.1 under win2k.
Another question that I'm sure is simple: is there a simple way to
find the list of
On Fri, 19 Mar 2004, Chris Evans wrote:
I prefer to use R on a linux box and ultimately need things to end up
there to serve things up using David Firth's excellent CGIwithR and
apache, but one step at a time and I've installed 1.8.1 under win2k.
Another question that I'm sure is simple:
Chris Evans wrote:
I prefer to use R on a linux box and ultimately need things to end up
there to serve things up using David Firth's excellent CGIwithR and
apache, but one step at a time and I've installed 1.8.1 under win2k.
Another question that I'm sure is simple: is there a simple way to
You should read ?daystoyears more carefully. There is a details section
where you read:
Details:
The years time-scale uses one unit for each year. We
deliberately linearized time in this time-scale and each year
has 365.25 days. There is thus no adjustment for bissextile years.
Hi,
I am interested in looking at cumulative density functions. If F(x) is a cumulative
density of monthly fund returns over the interval of a to b, and I am interested in
returns above and below a specified point r, then I would like to find the number that
is made up of
1.(integral from r
Wilhelm B. Kloke [EMAIL PROTECTED] writes:
I have been asked how to handle the following situation in R:
Given an unbalanced design of 3 crossed random effects, such as
subject, rater and item, how to estimate the variance components?
I know how to do it using lme, but this seems to be
Hi R users.
Why if I have a two-way or multifactor analysis of variance and
I call the function model.tables with the se=T options there
is not estimation when I have an interaction on the model?
Thank you very much for your help
--
Kenneth Cabrera
Universidad Nacional de Colombia
Tel Of 430
Dear list,
I've been messing around with coding functions in R and it just won't make
sense to me.
Running my analysis by hand on command line is fine and works but because
of the repetitive nature of the job I would like to code a function for it.
My problem:
I would like to read in data
Dear R-ers,
I am just a simple 'end-user' of R and am trying to analyse data with a binary
response variable (dead or alive) in relation to weight and sex (of young birds). As
some of the birds have the same biological mother, I am using mixed models with the
identity of the mother as a random
[This appears to have bounced so I am sending it again. Apologies
if it gets there twice.]
Thanks. Thus it seems that there are two types of expressions and calls:
1. fully expanded
2. partially expanded
and that fully expanded ones are a prerequisite for substitution.
body() and quote()
Because the result is a list, you can get your number like this:
integrate(dnorm,0,2)[[1]]
You don't have an expression for the probability function? You can do
a nonlinear regression to a distribution you belive in, and then
integrate that. Or, you can calculate the experimental cdf by simple
1. Section 2.6 Character Vectors of R-intro (www.r-project.org - Manual -
Introduction to R)
2. Search the archives https://www.stat.math.ethz.ch/mailman/listinfo/r-help
which is appended as footnote on every R-help mail
3. help(paste)
-Original Message-
From:
[EMAIL PROTECTED]
In an experimental setup we obtain z-data samples at equidistant grid points.
The surf.gls (Kriging) algorithm produces an error under this circumstance when
performing the Choleski decomposition.
A workaround is to dither the grid coordinates using (x - rnorm(length(x)) ; y-
On Fri, 19 Mar 2004 07:49:26 -0700
Neil Skjodt [EMAIL PROTECTED] wrote:
Hello:
I am new R user stumped why the R code after this paragraph generates
Error: syntax error messages after each of the last 2 lines. I have
tried searching the manuals, Hmisc documentation, contributed manuals,
On Fri, 19 Mar 2004, Stegemann, Berthold wrote:
In an experimental setup we obtain z-data samples at equidistant grid points.
The surf.gls (Kriging) algorithm produces an error under this
circumstance when performing the Choleski decomposition.
A workaround is to dither the grid
Try counting parentheses: they don't match. A good editor (e.g. R mode in
Emacs) would make this instantly obvious.
I doubt if data.df(subj,2:7) is what you want, either: perhaps
data.df[subj, 2:7]. There is much more in that vain.
On Fri, 19 Mar 2004, Neil Skjodt wrote:
Hello:
I am new
I can only assume I'm betraying my ignorance here, but this is not what
I would expect.
I'm getting the following from a stepwise selection (with both step and
stepAIC):
step(lm(sqrt(Grids)~ SE + Edge + NH), scope=~ (Edge + SE + NH)^2)
Start: AIC= 593.56
sqrt(Grids) ~ SE + Edge + NH
Hi R specialists,
When testing the association between two time series the cor.test gives
the following message...- p-values may be incorrect due to tie
What does it mean? (it is not described in the help)
Thankx,
Jan
cor.test(Origi[,1],Origi[,2], alternative = c(two.sided),method =
Try par(pty=s).
HTH,
Andy
From: Robin Hankin
Hi everyone
I want a square scatterplot with abline(0,1) going exactly through the
SW and NE corners. By square I mean that the plotting region is
exactly square, and that the axis limits are identical.
x - 1:20
y - x+rep(c(-1,1),10)
Robin Hankin wrote:
Hi everyone
I want a square scatterplot with abline(0,1) going exactly through the
SW and NE corners. By square I mean that the plotting region is
exactly square, and that the axis limits are identical.
x - 1:20
y - x+rep(c(-1,1),10)
lims - range(c(x,y))
None of the
plot(x,y,asp=1, axes=F, bty=n);abline(0,1) #diagonal line misses corners
axis(2, pos=0)
axis(1, pos=0)
hope this helps. spencer graves
Robin Hankin wrote:
Hi everyone
I want a square scatterplot with abline(0,1) going exactly through the
SW and NE corners. By square I mean that the
On Fri, 2004-03-19 at 09:06, Robin Hankin wrote:
Hi everyone
I want a square scatterplot with abline(0,1) going exactly through the
SW and NE corners. By square I mean that the plotting region is
exactly square, and that the axis limits are identical.
x - 1:20
y - x+rep(c(-1,1),10)
par(pty=s) is designed to set square plotting regions. (That is in `An
Introduction to R', as well as used frequently in MASS.)
plot(0:21, 0:21, type=n, xlab=x, ylab=x)
points(x,y)
abline(0,1)
is one way to do it thereafter, as well as
plot(x,y,xlim=lims,ylim=lims);abline(0,1)
On Fri, 19
Deepayan,
Both auto.key and simpleKey are convenience tools to make key drawing
easy in 'typical' cases. For more flexibility, define the key as a list
(the structure is described under 'key' in ?xyplot). You may want to
look at ?Rows as well, in conjunction with trellis.par.get().
See
Dear Neil,
In addition to Frank's observations, it's not a good idea to grow vectors
(here, control and test) in a loop. Doing so causes the vectors to be copied
over and over. While that doesn't make much difference for vectors with 11
elements, it can make a big difference in a large problem.
Anon. [EMAIL PROTECTED] writes:
I can only assume I'm betraying my ignorance here, but this is not
what I would expect.
I'm getting the following from a stepwise selection (with both step
and stepAIC):
step(lm(sqrt(Grids)~ SE + Edge + NH), scope=~ (Edge + SE + NH)^2)
Start: AIC=
On Fri, 2004-03-19 at 09:29, Marc Schwartz wrote:
On Fri, 2004-03-19 at 09:06, Robin Hankin wrote:
Hi everyone
I want a square scatterplot with abline(0,1) going exactly through the
SW and NE corners. By square I mean that the plotting region is
exactly square, and that the axis
Jan Verbesselt [EMAIL PROTECTED] writes:
Hi R specialists,
When testing the association between two time series the cor.test gives
the following message...- p-values may be incorrect due to tie
What does it mean? (it is not described in the help)
It means what it says... The p-values in
Begin forwarded message:
From: Dan Kelley [EMAIL PROTECTED]
Date: March 19, 2004 12:00:01 PM AST
To: Don MacQueen [EMAIL PROTECTED]
Subject: Re: [R] osx/fink: cannot do R INSTALL (library mixup)
That works perfectly! THanks. I did
524 export PKG_LIBS=-L/usr/local/lib -L/sw/lib
525 R
Hi all,
I created a data frame with three factors, plus the response that
looks like this:
x1 x2 x3 y
a 1 1 0.3
a 2 1 0.1
b 1 1 0.4
c 4 3 0.1
...
I would like to analise the effect of two of them, keeping the
Hi All,
I have use the BLAST program provided by NCBI to compare large databases and have
generated a lot of output in one long text file. I was looking for a way to parse
out the results in a nice way so that it could be easily visualized. I was thinking
of trying to write something in R, but
Hi,
Quick question on reading data.
I am running simulations but want to allow the user the option to define the number of
simulations. How can I have R read-in user data entered from the keyboard? Is there a
difference for reading in numeric and character data?
For example, I am trying to
On Fri, 2004-03-19 at 10:47, Kissell, Robert [EQRE] wrote:
Hi,
Quick question on reading data.
I am running simulations but want to allow the user the option to
define the number of simulations. How can I have R read-in user data
entered from the keyboard? Is there a difference for
Giampiero Salvi wrote:
I created a data frame with three factors, plus the response that
looks like this:
x1 x2 x3 y
a 1 1 0.3
a 2 1 0.1
b 1 1 0.4
c 4 3 0.1
...
I would like to analise the effect of two of them,
On Fri, Mar 19, 2004 at 11:03:07AM -, Chris Evans wrote:
This has probably been addressed here, but I can't find it if it has,
and it may be up on the R project homepage or the CRAN pages, but if
so
... so sorry if this has been answered before: question is, can I
upgrade to
Giampiero,
It's certainly possible.
Many commands allow the subset option, which permits the analysis of only a
portion of a dataframe, although boxplot does not.
Also, you could preface your analysis with a logical condition, like
my sample - x3 == 1
boxplot(y[my sample] ~ x1[my sample]
I would like to collect the trees grown by rpart fits in an array,
in order to be able to use them later to predict new data. I have
tried to use parse and eval to do this, without success. I'd be
very grateful if someone could explain how to do it.
The kind of thing I hope to do is:
On Fri, 19 Mar 2004, Giampiero Salvi wrote:
Hi all,
I created a data frame with three factors, plus the response that
looks like this:
x1x2 x3 y
a 1 1 0.3
a 2 1 0.1
b 1 1 0.4
c 4 3 0.1
...
I would like to
Jonathan,
Try making a list instead of an array. See ?list. Also, did you look into
random forests? I'm not sure what you want to do, but there might be
methods there to do some of the work for you.
Sean
On 3/19/04 1:12 PM, Jonathan Williams
[EMAIL PROTECTED] wrote:
I would like to collect
How to use contour function?
I type contour(th1,th2,SumofSquares,levels=c(seq(1000,4000,200)))
But I didn't see contour in the plot, I guess it is because my data is so
big that contour is out of range.
Does anybody know how to fix this problem?
here th1 is a 1x101 array(158741.8-298529.6)
th2
On Fri, 19 Mar 2004 11:46:25 -0500 (EST), Matthew Oliver
[EMAIL PROTECTED] wrote :
Hi All,
I have use the BLAST program provided by NCBI to compare large databases and have
generated a lot of output in one long text file. I was looking for a way to parse
out the results in a nice way so that
Yes, Jonathan has looked at randomForest, but for rpart trees, ipred might
be more appropriate.
Cheers,
Andy
From: Sean Davis
Jonathan,
Try making a list instead of an array. See ?list. Also, did
you look into
random forests? I'm not sure what you want to do, but there might be
On Fri, 19 Mar 2004 11:47:10 -0500, Kissell, Robert [EQRE]
[EMAIL PROTECTED] wrote :
Hi,
Quick question on reading data.
I am running simulations but want to allow the user the option to define the number
of simulations. How can I have R read-in user data entered from the keyboard? Is
there a
Matt -
Yes, you would be seriously re-inventing the wheel.
A standard tool is the bioperl SearchIO package.
This is described at
www.bioperl.org Docs SearchIO HOWTO:
The html version of that link evaluates to:
http://bioperl.org/HOWTOs/html/SearchIO.html
- best - tom blackwell - u
Jonathan Williams wrote:
I would like to collect the trees grown by rpart fits in an array,
in order to be able to use them later to predict new data. I have
tried to use parse and eval to do this, without success. I'd be
very grateful if someone could explain how to do it.
The kind of thing I
Aimin Yan wrote:
How to use contour function?
I type contour(th1,th2,SumofSquares,levels=c(seq(1000,4000,200)))
But I didn't see contour in the plot, I guess it is because my data is
so big that contour is out of range.
It looks like it, given the data you showed. If you leave out the
'levels'
Aside from the console solution already mentioned there are a few others:
- If you are using Windows you can use the winDialog function.
- If you want to have the user select from a fixed number of alternatives you
can use the select.list function.
- you can use tcl/tk to create the user
The goal: simulate chi square mixture distributions as a way of
simulating likelihood ratio test statistics for some mixed models where
the more specific model has some zero variance components (a la Pinheiro
and Bates pg. 84-87)
The problem: R doesn't have the function ms which is apparently
The following response by B.Ripley to a similar request may help.
Alex
On Tue, 21 Oct 2003, Ernie Adorio wrote:
Am using R on a Linux box and am currently writing an interactive R
script.
1. How do I ask a user to press any key to continue ? I used a system call
to
read but this only
My appologies to everyone.
Prof. Ripley, please realize that I did not intend to suggest that your
mathematics were wrong. My question about mathematical correctness
refered to my presention of your formula as an equation. I have great
admiration for your work and its influences on my field and,
Scott Rifkin [EMAIL PROTECTED] writes:
The goal: simulate chi square mixture distributions as a way of
simulating likelihood ratio test statistics for some mixed models where
the more specific model has some zero variance components (a la Pinheiro
and Bates pg. 84-87)
The problem: R
One alternative can be found in my under development (slow, and
probably won't be finished) BioSeq1 package; it can be gotten
painfully via:
http://www.bioconductor.org/develPage/develPackages.html
There are tools for getting BLAST results back and processing them.
best,
-tony
Tom
Hello
I have data in many files in a directory, how can I
loop through the files in a given dir in-order-to
build a data.frame?
thanks
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting
d - dir(c:/data)
for (i in d){
dt - read.csv(c(c:/data/,i),header=FALSE)
}
wouldn't work because c(c:/data,i) puts out string
string where both strings need to be one string.
how can I combinde both in one string, c:/data/i like
the good old perl I wish.
thanks
Hi,
I want to use R_qsort_int_I() in my C function, but getting the
following error. It looks like there is a conflict between Rmath.h,
which I use to generate random numbers, and R_ext/Boolean.h I would
appreciate any help to fix this problem.
gcc -ansi -g -o Gibbs gibbs.c subroutines.o
Fred J. [EMAIL PROTECTED] writes:
d - dir(c:/data)
for (i in d){
dt - read.csv(c(c:/data/,i),header=FALSE)
}
wouldn't work because c(c:/data,i) puts out string
string where both strings need to be one string.
how can I combinde both in one string, c:/data/i like
the good old perl I
I have a microarray analysis program using Monte Carlo Markov chain. I
downloaded R 1.9 alpha and saw FOUR times increase in speed.
Completelyly amazing. Specifically, one iteration takes 44 seconds
under R 1.8.1 and only 10 seconds under R 1.9.0 alpha. I have averaged
over many iterations so
file.path and paste.
thanks
as a R starter, how could I have found out without
posting a question. is there a way I could have
searched the help files for a keyword or somthing?
thanks
__
[EMAIL PROTECTED] mailing list
Hello
I have a data frame with many col.s and rows
V4 V5 V6 V7
3 4 5 6
1 4 4 1
2 4 4 1
4 0 5 1
since the data has the middle 2 rows with V5 and V6
are equal, I need to produce
V4 V5 V6 V7
3 4 5 6
this line removed and the value V7 = 1 is
added to
help.search(string)
or
help.search(concatenate)
would probably have led you to it.
Fred J. wrote:
file.path and paste.
thanks
as a R starter, how could I have found out without
posting a question. is there a way I could have
searched the help files for a keyword or somthing?
thanks
Hello again,
I got three responses for help on the leading zero problem. Thank you.
Alas I still don't have it working. Here are more specifics:
I read in a data file like this:
participants-read.table(C:/Work/blah-blah)
The data file consists of the fields last name, first name, social
Great detective work, Tony! Thanks.
Date: Fri, 19 Mar 2004 17:51:27 -0700
From: Tony Plate [EMAIL PROTECTED]
To: [EMAIL PROTECTED], [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Subject: Re: [R] substitute question
I don't think it is correct that there are two types of expressions
There seems to be some obscure features in step() when you have
interaction terms: their interpretation is order sensitive. Term A:B is
regarded to be something else than B:A. Sometimes this results in error
Error in factor.scope(ffac, list(add = fadd, drop = fdrop)) :
upper scope does
On Fri, 19 Mar 2004, Randy Zelick wrote:
I got three responses for help on the leading zero problem. Thank you.
Well, it seems that you didn't tell us what the actual problem was: please
consult the posting guide and its references and learn to ask the right
question.
Alas I still don't have
82 matches
Mail list logo