Re: [R] use factor for categorical covariate in Cox PH model
Linda Lei llei at bccrc.ca writes: When dealing with the categorical covariates (for example 3 groups), it will come out different results if we add the command factor in front of the categorical covariate or not The catch is here: the covariate is not a categorial, but a number. In that case a regression is computed, and the ONE ration printed out has the meaning of a slope. if we don't add factor, there is only one hazard ratio; if we add factor, there are two hazard ratios. So does the factor actually create some dummy variables for the calculation? Factor make the covariate a categorial one, and two contrasts are printed out, relative to the first level of the covariate. Dieter __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Output and Word
Hi I have just started to have a look at R. I have used most stats software packages and can use perl, visual basic etc. I am interested in how well it handles lots of output e.g. tables or charts. How would you get lots of output most easily and quickly into a Word document? Sharon Snowdon - FIGHT BACK AGAINST SPAM! Download Spam Inspector, the Award Winning Anti-Spam Filter http://mail.giantcompany.com -- 7/14/2006 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] String manipulation and formatting
Thanks to Richard, Gabor and Marc for some nice solutions to my request.
I have a new problem:
xify(30.10)
[1] X.X
xify(30.11)
[1] XXX.XXX
The problem originates from:
as.numeric(unlist(strsplit(as.character(15.10), \\.)))
[1] 15 1
as.numeric(unlist(strsplit(as.character(15.11), \\.)))
[1] 15 11
It seems to boils down to:
as.character(15.10)
[1] 15.1
A simple solution is:
xify(15.10)
[1] X.XX
I was wondering if there is a simple way for xify to see the zero of the 10
without having to force the user to add quotes around the format?
Thanks,
Saghir
-Original Message-
From: Marc Schwartz (via MN) [mailto:[EMAIL PROTECTED]
Sent: Monday, July 17, 2006 16:55
To: Bashir Saghir (Aztek Global)
Cc: '[EMAIL PROTECTED]'
Subject: Re: [R] String manipulation and formatting
snip
Here are two variations:
xify - function(x)
{
exxes - as.numeric(unlist(strsplit(as.character(x), \\.)))
ifelse(length(exxes) == 2,
paste(paste(rep(X, exxes[1] - exxes[2]), collapse = ),
paste(rep(X, exxes[2]), collapse = ),
sep = .),
paste(rep(X, exxes[1]), collapse = ))
}
xify - function(x)
{
exxes - as.numeric(unlist(strsplit(as.character(x), \\.)))
tmp - sapply(exxes, function(x) paste(rep(X, x), collapse = ))
ifelse(length(tmp) == 2,
paste(substr(tmp[1], 1, exxes[1] - exxes[2]), tmp[2],
sep = .),
tmp)
}
HTH,
Marc Schwartz
-Original Message-
From: Richard M. Heiberger [mailto:[EMAIL PROTECTED]
Sent: Monday, July 17, 2006 16:42
To: Bashir Saghir (Aztek Global); '[EMAIL PROTECTED]'
Subject: Re: [R] String manipulation and formatting
xify - function(number) {
fn - format(number)
fn3 - unlist(strsplit(fn, \\.))
if (length(fn3) == 1) paste(rep(X, as.numeric(fn3)), collapse=)
else
paste(paste(rep(X, as.numeric(fn3)[1]-as.numeric(fn3)[2]), collapse=),
paste(rep(X, as.numeric(fn3)[2]), collapse=),
sep=.)
}
-
Legal Notice: This electronic mail and its attachments are i...{{dropped}}
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] String manipulation and formatting
Hi,
an other way without any libraries and written as a one-line-command
would be something like this:
xify - function(x)
{
gsub([0-9],X,
sprintf(
paste(
%,
ifelse(
format(x,digit=1)==as.character(x),
paste(as.character(x),.0,sep=),
as.character(x)
),
f,sep=
)
,10^(trunc(x)-10*x+10*trunc(x)-1)
)
)
}
The only disadvantage is that something like xify(4.8) produces [1]
X..
Hans-Joerg
On 7/17/06, Bashir Saghir (Aztek Global) [EMAIL PROTECTED]
group.com wrote:
I'm trying to write a simple function that does the following:
[command] xify(5.2)
[output] XXX.XX
[command] xify(3)
[output] XXX
Any simple solutions (without using python/perl/unix script/...)?
Thanks,
Saghir
-
Legal Notice: This electronic mail and its attachments are i...
{{dropped}}
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/
posting-guide.html
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-
guide.html
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] String manipulation and formatting
On Tue, 18 Jul 2006, Bashir Saghir (Aztek Global) wrote: Thanks to Richard, Gabor and Marc for some nice solutions to my request. I have a new problem: xify(30.10) [1] X.X xify(30.11) [1] XXX.XXX The problem originates from: as.numeric(unlist(strsplit(as.character(15.10), \\.))) [1] 15 1 as.numeric(unlist(strsplit(as.character(15.11), \\.))) [1] 15 11 It seems to boils down to: as.character(15.10) [1] 15.1 A simple solution is: xify(15.10) [1] X.XX I was wondering if there is a simple way for xify to see the zero of the 10 without having to force the user to add quotes around the format? No, as the parser has converted '15.10' to a numeric constant. What is the problem with asking users to enter strings as strings? -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Nested functions
Am Montag, den 17.07.2006, 19:42 -0400 schrieb jim holtman:
You were down at least 5300 levels in subroutine calls. Here is the
first couple of lines from 'traceback()':
5363: vavb(a, b, n, v, x)
...
5316: vavb(a, b, n, v, x)
The other thing is that there is no 'x' defined in the function;
antiphonar - function(v)
{
a - 1;
b - 2;
n - length(v);
alessn(a, b, n, v, x);
}
and this is the initial call to the rest of the functions.
Ooops... normally I don't program THAT bad ... ;-)
What is the problem you are trying to solve?
I'm analyzing a 16th century music manuscript. There is no rhythm
notated there, but I can to a certain degree restore it by finding
self-similarities in the melodic structure. The vector v is an ordered
list of intervals. I could find them by hand, but this takes time and
things are easily missed. Besides, for a 200 pages manuscript coding the
vectors takes long, but finding similarities by hand takes
endless... :-(
The problem remains. I'm not sure I did completely understand the way R
works. My plan was to jump out of one function to another. I think what
R does is assigning a function's return value to an internal variable
and tries to evaluate another function WITHIN a calling function, which
of course leads to deep nesting and a bunch of internal variables. Too
bad there is no goto statement in R... Isn't there a way to make R
forget that it just evaluated e.g. alessn the moment it calles vavb?
That algorithm itself isn't that hard. In the end it's just a decision
diagram of the form if THIS, goto THIS knot, otherwise goto THAT knot.
Each knot has only one arrow pointing away from it, but sometimes more
than one pointing toward it. I could send the diagram as an attachement, if you
like.
Cheers,
John
P.S.: Please keep messages on the list. Thank you!
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] String manipulation and formatting
Does it have to be a stop char ., or could it be a separate parameter, i.e. put in a comma, then the 10 becomes an integer... s/ On 18/07/06, Prof Brian Ripley [EMAIL PROTECTED] wrote: On Tue, 18 Jul 2006, Bashir Saghir (Aztek Global) wrote: Thanks to Richard, Gabor and Marc for some nice solutions to my request. I have a new problem: xify(30.10) [1] X.X xify(30.11) [1] XXX.XXX The problem originates from: as.numeric(unlist(strsplit(as.character(15.10), \\.))) [1] 15 1 as.numeric(unlist(strsplit(as.character(15.11), \\.))) [1] 15 11 It seems to boils down to: as.character(15.10) [1] 15.1 A simple solution is: xify(15.10) [1] X.XX I was wondering if there is a simple way for xify to see the zero of the 10 without having to force the user to add quotes around the format? No, as the parser has converted '15.10' to a numeric constant. What is the problem with asking users to enter strings as strings? -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Object name and Strings?
Hi all, Is there a simple way to convert an object name to a characters string? Stéphane. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Object name and Strings?
deparse(substitute(a)) [1] a --- Jacques VESLOT CNRS UMR 8090 I.B.L (2ème étage) 1 rue du Professeur Calmette B.P. 245 59019 Lille Cedex Tel : 33 (0)3.20.87.10.44 Fax : 33 (0)3.20.87.10.31 http://www-good.ibl.fr --- Stéphane Cruveiller a écrit : Hi all, Is there a simple way to convert an object name to a characters string? Stéphane. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] String manipulation and formatting
In my opinion there is nothing wrong with asking the user to enter strings
as strings. I am working on some functions for SAS users (with limited R/S
experience) who are used to working with the SAS macro language. I was
trying to keep things simple for them by not forcing the use of quotes (a
style they use with their SAS macros). Given your explanation I'll just
document this feature in the help pages and use my time more productively
else where.
Thanks,
Saghir
-Original Message-
From: Prof Brian Ripley [mailto:[EMAIL PROTECTED]
Sent: Tuesday, July 18, 2006 10:24
To: Bashir Saghir (Aztek Global)
Cc: '[EMAIL PROTECTED]'
Subject: Re: [R] String manipulation and formatting
On Tue, 18 Jul 2006, Bashir Saghir (Aztek Global) wrote:
Thanks to Richard, Gabor and Marc for some nice solutions to my request.
I have a new problem:
xify(30.10)
[1] X.X
xify(30.11)
[1] XXX.XXX
The problem originates from:
as.numeric(unlist(strsplit(as.character(15.10), \\.)))
[1] 15 1
as.numeric(unlist(strsplit(as.character(15.11), \\.)))
[1] 15 11
It seems to boils down to:
as.character(15.10)
[1] 15.1
A simple solution is:
xify(15.10)
[1] X.XX
I was wondering if there is a simple way for xify to see the zero of the
10
without having to force the user to add quotes around the format?
No, as the parser has converted '15.10' to a numeric constant.
What is the problem with asking users to enter strings as strings?
--
Brian D. Ripley, [EMAIL PROTECTED]
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax: +44 1865 272595
-
Legal Notice: This electronic mail and its attachments are i...{{dropped}}
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] String manipulation and formatting
As I am trying to maintain consistency with some SAS macros I prefer not to
change the standard definitions in use at the moment. If I change things
too much I don't think many people with use my new functions.
Thanks,
Saghir
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of Sean O'Riordain
Sent: Tuesday, July 18, 2006 10:50
To: Bashir Saghir (Aztek Global)
Cc: [EMAIL PROTECTED]
Subject: Re: [R] String manipulation and formatting
Does it have to be a stop char ., or could it be a separate
parameter, i.e. put in a comma, then the 10 becomes an integer...
s/
On 18/07/06, Prof Brian Ripley [EMAIL PROTECTED] wrote:
On Tue, 18 Jul 2006, Bashir Saghir (Aztek Global) wrote:
Thanks to Richard, Gabor and Marc for some nice solutions to my request.
I have a new problem:
xify(30.10)
[1] X.X
xify(30.11)
[1] XXX.XXX
The problem originates from:
as.numeric(unlist(strsplit(as.character(15.10), \\.)))
[1] 15 1
as.numeric(unlist(strsplit(as.character(15.11), \\.)))
[1] 15 11
It seems to boils down to:
as.character(15.10)
[1] 15.1
A simple solution is:
xify(15.10)
[1] X.XX
I was wondering if there is a simple way for xify to see the zero of the
10
without having to force the user to add quotes around the format?
No, as the parser has converted '15.10' to a numeric constant.
What is the problem with asking users to enter strings as strings?
--
Brian D. Ripley, [EMAIL PROTECTED]
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax: +44 1865 272595
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
-
Legal Notice: This electronic mail and its attachments are i...{{dropped}}
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Object name and Strings?
On Tue, 18 Jul 2006, Stéphane Cruveiller wrote: Hi all, Is there a simple way to convert an object name to a characters string? Yes, as.character, as in x - as.name(foo) x foo as.character(x) [1] foo However, I suspect you are not using the words in their technical sense (a name is a synonym for a symbol in R), so if this is not the answer, please give us an example of what you are trying to do (which might be deparse). -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595__ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] package installation problems
Hello, I just updated from R 2.1.1 to R 2.3.1 (I also updated from Fedora 3 to Fedora 4 if that is of any importance). However, several packages (e.g. fields etc.) refuse to be installed. I get the following error message: [EMAIL PROTECTED] R-files]# R CMD INSTALL fields_2.3.tar.gz * Installing *source* package 'fields' ... ** libs g77 -fpic -O2 -g -pipe -m32 -march=i386 -mtune=pentium4 -c css.f -o css.o f771: invalid option `tune=pentium4' make: *** [css.o] Error 1 ERROR: compilation failed for package 'fields' ** Removing '/usr/lib/R/library/fields' For some reason -mtune is set to Pentium 4, although I'm using an Athlon-XP. I searched the files in the fields package for this line but couldn't find it. Other packages, like f.i. maps didn't have any problems. I hope anyone can help me, as I need those packages urgently for my PhD project. Joanne __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Large datasets in R
Il giorno lun, 17/07/2006 alle 15.00 -0400, Deepankar Basu ha scritto: I have been trying to read up the posting on the R-archive on this topic; but I could not really understand all the discussion, nor could I reach the end. So, I am not aware of the current state of consensus on the issue. a general hint is to store this dataset in a database and manage your data with RODBC or other db-related packages. cheers -- Daniele Medri __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Object name and Strings?
Thanks for your answer. Here is what I am trying to do. I have a list which is called MyList. I would like to get only the name of this object as a simple characters string. i.e. Is there a function in R which is able to give: name-fun(Mylist) name MyList thanks in advance, Stéphane. Prof Brian Ripley a écrit : On Tue, 18 Jul 2006, Stéphane Cruveiller wrote: Hi all, Is there a simple way to convert an object name to a characters string? Yes, as.character, as in x - as.name(foo) x foo as.character(x) [1] foo However, I suspect you are not using the words in their technical sense (a name is a synonym for a symbol in R), so if this is not the answer, please give us an example of what you are trying to do (which might be deparse). __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Object name and Strings?
On Tue, 18 Jul 2006, Stéphane Cruveiller wrote: Thanks for your answer. Here is what I am trying to do. I have a list which is called MyList. I would like to get only the name of this object as a simple characters string. i.e. Is there a function in R which is able to give: name-fun(Mylist) name MyList myfun - function(x) deparse(substitute(x)) That's how to find the name (if it was a name) given as a function argument. thanks in advance, Stéphane. Prof Brian Ripley a écrit : On Tue, 18 Jul 2006, Stéphane Cruveiller wrote: Hi all, Is there a simple way to convert an object name to a characters string? Yes, as.character, as in x - as.name(foo) x foo as.character(x) [1] foo However, I suspect you are not using the words in their technical sense (a name is a synonym for a symbol in R), so if this is not the answer, please give us an example of what you are trying to do (which might be deparse). -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595__ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Object name and Strings?
That is exactly what I wanted to do... Thanks for the hint... Stéphane. Prof Brian Ripley a écrit : On Tue, 18 Jul 2006, Stéphane Cruveiller wrote: Thanks for your answer. Here is what I am trying to do. I have a list which is called MyList. I would like to get only the name of this object as a simple characters string. i.e. Is there a function in R which is able to give: name-fun(Mylist) name MyList myfun - function(x) deparse(substitute(x)) That's how to find the name (if it was a name) given as a function argument. thanks in advance, Stéphane. Prof Brian Ripley a écrit : On Tue, 18 Jul 2006, Stéphane Cruveiller wrote: Hi all, Is there a simple way to convert an object name to a characters string? Yes, as.character, as in x - as.name(foo) x foo as.character(x) [1] foo However, I suspect you are not using the words in their technical sense (a name is a synonym for a symbol in R), so if this is not the answer, please give us an example of what you are trying to do (which might be deparse). __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Spectral data analysis - parameter selection
Hello all, I am doing some spectral data analysis using R, and I already got some help on how to do savitsky-golay preprocessing. Now I would like to perform parameter selection (i.e. I would like to assess which spectral lines are relevant to keep in the model, and which are not). There exist algorithms like the jacknife algorithm, but I was wondering whether anyone has already implemented something like that, and if not, if anyone could give me a few hints on how to start implementing it myself. Thanks in advance, Dirk -- Dirk De Becker Work: Kasteelpark Arenberg 30 3001 Heverlee phone: ++32(0)16/32.14.44 fax: ++32(0)16/32.85.90 Home: Waversebaan 90 3001 Heverlee phone: ++32(0)16/23.36.65 [EMAIL PROTECTED] mobile phone: ++32(0)498/51.19.86 Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] String manipulation and formatting
Hi,
only if you allow to input x as a string then you can use, maybe,
simply the following one-line-command:
xify - function(x)
{
gsub(
[0-9],
X,
sprintf(
paste(%,ifelse(regexpr(\\.,x) 0, x, x-paste(x,.
0,sep=)), f,sep=),
10^(trunc(as.numeric(x))-as.integer(gsub((.*?)\\.(.*?),\\2,x))-1)
)
)
}
xify(30.10)
[1] .XX
Hans-Joerg
Thanks to Richard, Gabor and Marc for some nice solutions to my
request.
I have a new problem:
xify(30.10)
[1] X.X
xify(30.11)
[1] XXX.XXX
The problem originates from:
as.numeric(unlist(strsplit(as.character(15.10), \\.)))
[1] 15 1
as.numeric(unlist(strsplit(as.character(15.11), \\.)))
[1] 15 11
It seems to boils down to:
as.character(15.10)
[1] 15.1
A simple solution is:
xify(15.10)
[1] X.XX
I was wondering if there is a simple way for xify to see the zero
of the
10
without having to force the user to add quotes around the format?
No, as the parser has converted '15.10' to a numeric constant.
What is the problem with asking users to enter strings as strings?
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] package installation problems
On Tue, 18 Jul 2006, Demmler J. wrote: Hello, I just updated from R 2.1.1 to R 2.3.1 (I also updated from Fedora 3 to How? Did you install an RPM? Before or after OS update? I checked the RPM for fc4 on CRAN which has C compiler:gcc -O2 -g -pipe -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m32 -march=i386 -mtune=pentium4 -fasynchronous-unwind-tables Fortran 77 compiler: gfortran -O2 -g -pipe -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m32 -march=i386 -mtune=pentium4 -fasynchronous-unwind-tables C++ compiler: g++ -O2 -g -pipe -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m32 -march=i386 -mtune=pentium4 -fasynchronous-unwind-tables Fortran 90/95 compiler:gfortran -g -O2 not as you show. However, read on. Fedora 4 if that is of any importance). However, several packages (e.g. fields etc.) refuse to be installed. I get the following error message: [EMAIL PROTECTED] R-files]# R CMD INSTALL fields_2.3.tar.gz * Installing *source* package 'fields' ... ** libs g77 -fpic -O2 -g -pipe -m32 -march=i386 -mtune=pentium4 -c css.f -o css.o f771: invalid option `tune=pentium4' make: *** [css.o] Error 1 ERROR: compilation failed for package 'fields' ** Removing '/usr/lib/R/library/fields' For some reason -mtune is set to Pentium 4, although I'm using an Athlon-XP. That may not actually be a problem: the problem is that your compiler is not recognizing -mtune. (I seem to have run R RPMs for fc3 with -march=i386 -mtune=pentium4 on a dual Athlon MP. 'tune' should generate code that runs on other CPUs, just not optimally.) I searched the files in the fields package for this line but couldn't find it. Other packages, like f.i. maps didn't have any problems. I hope anyone can help me, as I need those packages urgently for my PhD project. The options are coming from the version of R you installed. I would expect R compiled for FC4 to be using gfortran not g77 (as in the CRAN RPM). The simplest way out here is compile R from the sources on your actual machine, but you could try replacing your RPM with the CRAN fc4 one. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Output and Word
Hi, you might try to use the R2HTML package and then import the html files into word. please see. http://cran.r-project.org/doc/Rnews/Rnews_2003-3.pdf Another way: use sweave for tex/latex output and transform it to rtf. Friedrich Schuster --- Your post was: sharon snowdon sharonsnowdon at fastmail.fm Hi I have just started to have a look at R. I have used most stats software packages and can use perl, visual basic etc. I am interested in how well it handles lots of output e.g. tables or charts. How would you get lots of output most easily and quickly into a Word document? Sharon Snowdon __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Nested functions
I was checking out your function and thinking of trying to make it more efficient when I noticed your comment above. :-D Yes, bioinformatics is full of this stuff - have a look at Bioconductor. I'm about it, but it'll take some time. It's a vast repository... For the meantime, I've attached a graph for clarification ;-) Dashed lines are followed if the test returns FALSE, solid if TRUE. Cheers, John diagramm.ps Description: PostScript document __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Running R as root
Hello Using R v. 2.3.1 which I installed recently, when I start it as root, there is nothing I can do. I get Error: function xx not found even when trying to leave R using q(). R works fine when started by an ordinary user. I rechecked the root environment several times and even have set it similar to that of an ordinary user (values of PATH etc.) but nothing helps. Using the command env did not reveal anything valuable. Going to root through the identity of an ordinary user and retaining his environment by doing su user instead of su - user does not help either. This does not seem related to the particular version of R. Using: Linux RedHat AS 2.1 Any hint? Bye Rick __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Running R as root
[EMAIL PROTECTED] writes: Hello Using R v. 2.3.1 which I installed recently, when I start it as root, there is nothing I can do. I get Error: function xx not found even when trying to leave R using q(). R works fine when started by an ordinary user. I rechecked the root environment several times and even have set it similar to that of an ordinary user (values of PATH etc.) but nothing helps. Using the command env did not reveal anything valuable. Going to root through the identity of an ordinary user and retaining his environment by doing su user instead of su - user does not help either. This does not seem related to the particular version of R. Using: Linux RedHat AS 2.1 Any hint? File permissions? (it is possible to set them so that files can be read by anyone except the user...) How exactly did you install R, and are there any other clues upon startup? I get, BTW aa() Error: could not find function aa so I suspect you paraphrased the error message. One other thing that can cause trouble is networked directories. Files that are physically on another machine can sometimes not be accessed by root on the local machine. -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Package for autocorrelation analysis?
Hi! Dear All, I want to implement an autocorrelation analysis and estimated the Moran's I in a set of ecological traits. I seek for an R package that do this analysis, however, I couldn't found none that implement it. May be I'm lost in the universe of the Contributed Packages but I couldn't found information into the index of some packages that I suppose do it. Would you give me some recommendations to do an autocorrelation analysis? Thanks! Ivan Ivan V. Rubio Pérez [EMAIL PROTECTED] Instituto de Ecología, UNAM www.ecologia.unam.mx -- Open WebMail Project (http://openwebmail.org) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Running R as root
Problem solved: I had a .Renviron file in root's Home directory pointing to the wrong R installation... Thanks anyway Rick __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Output and Word
If its text that you want included in a larger Word document you can create a text or HTML file from R and then when inserting it into Word insert it as a link rather than copying it in. output a file from R In Word Insert | File browse to the file so its name appears in the File name box click on the down arrow to the right of the Insert button and choose Insert as Link If you regenerate the file in the future your Word document gets automatically uipdated. Another possibility is either the rcom and RDCOMClient packages which allow one to control Word from R by treating Word as a COM object. This allows detailed control of Word but is more work. e.g. Here we create a Word document from scratch. Note that you must be running Windows and have Word on the machine where you do this: library(RDCOMClient) ow - COMCreate(Word.Application) ow[[Visible]] - TRUE # optional od - ow[[Documents]]$Add() od[[PageSetup]][[LeftMargin]] - 20 od[[PageSetup]][[TopMargin]] - 20 od[[Content]][[Font]][[Size]] - 11 od[[Content]][[Text]] - Hello World od$SaveAs(\\mydoc.doc) ow$Quit() On 7/17/06, sharon snowdon [EMAIL PROTECTED] wrote: Hi I have just started to have a look at R. I have used most stats software packages and can use perl, visual basic etc. I am interested in how well it handles lots of output e.g. tables or charts. How would you get lots of output most easily and quickly into a Word document? Sharon Snowdon - FIGHT BACK AGAINST SPAM! Download Spam Inspector, the Award Winning Anti-Spam Filter http://mail.giantcompany.com -- 7/14/2006 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Surv analysis with multiple internal time-dep covariates measured over different time intervals
Hi, I am analysing survival data (diagnosis time until death/cens) with time-dependent covariates. I would like to fit a cox model using the (start, stop] variable. In summary, I have the multiple internal time dependent covariates as follows; 1). LAS score (measured weekly on low risk patients, monthly on high risk) 2). EORTC score (measured monthly on low risk patients and every 3 months on high risk) 3). BMI (measured monthly on low risk patients and every 3 months on high risk) I have referred to the John Fox 'Cox Proportional-Hazards Regression for Survival Data' http://cran.r-project.org/doc/contrib/Fox-Companion/appendix-cox-regression.pdf and the corresponding script file at http://cran.r-project.org/doc/contrib/Fox-Companion/cox-regression.txt and also to Therneau and Grambsch. My problem is creating the dataset, possibly using the fold function (as described in Fox, p9) with more than one time-dependent covariate (which I successfully did with LAS). I have longitudinal measurements for each subject (with each date of assessment) as above with some missing data in a period of time before death (which I have entered as NA). Since the measurements in time depend on whether the patient is high or low risk and are made at different time intervals for each covariate, I wasn't sure how to code this in R. I would be really grateful if anyone could start me off on this. Thank you to those who respond. Zoe __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Output and Word
R can handle as many graphs/table as you ask it to. I don't know what you're trying to do, but I generate lots of graphs using loops and generating filenames on the spot. See FAQ 7.34 to see how it's done. http://cran.r-project.org/doc/FAQ/R-FAQ.html#How-can-I-save-the-result-of-each-iteration-in-a-loop-into-a-separate-file_003f For graphs, save the graphs as something windows can recognize (jpeg, png, emf, wmf...) and in Word, you can easily import all the graphs in a directory by doing Insert/Picture/From File and shift-click all the graphs you want. Word will import them all. For tables, it really depends on what you want because usually there is formatting involved. Personnaly, if I had to produce several tables to be imported in Word, I would use the R2HTML library in a loop (as mentionned above) and import the multiple html files using Insert/File (and shift-click) in Word. Neuro From: sharon snowdon [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Subject: [R] Output and Word Date: Tue, 18 Jul 2006 06:35:38 +1000 Hi I have just started to have a look at R. I have used most stats software packages and can use perl, visual basic etc. I am interested in how well it handles lots of output e.g. tables or charts. How would you get lots of output most easily and quickly into a Word document? Sharon Snowdon - FIGHT BACK AGAINST SPAM! Download Spam Inspector, the Award Winning Anti-Spam Filter http://mail.giantcompany.com -- 7/14/2006 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html From: sharon snowdon [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Subject: [R] Output and Word Date: Tue, 18 Jul 2006 06:35:38 +1000 Hi I have just started to have a look at R. I have used most stats software packages and can use perl, visual basic etc. I am interested in how well it handles lots of output e.g. tables or charts. How would you get lots of output most easily and quickly into a Word document? Sharon Snowdon - FIGHT BACK AGAINST SPAM! Download Spam Inspector, the Award Winning Anti-Spam Filter http://mail.giantcompany.com -- 7/14/2006 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Nested functions
Please look at http://www.turbulence.org/Works/song/ This is a website by Martin Wattenberg that visually displays the types of patterns you are looking for. He gave a paper at the Joint Statistics Meetings in Minneapolis 2005. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Package for autocorrelation analysis?
On Tue, 2006-07-18 at 07:10 -0500, Ivan Rubio Perez wrote: Hi! Dear All, I want to implement an autocorrelation analysis and estimated the Moran's I in a set of ecological traits. I seek for an R package that do this analysis, however, I couldn't found none that implement it. May be I'm lost in the universe of the Contributed Packages but I couldn't found information into the index of some packages that I suppose do it. Would you give me some recommendations to do an autocorrelation analysis? Thanks! Ivan Please, learn to help yourself and use the tools provided. Also, don't post a new message to the list by replying to an existing thread it really messes up the archives! Try, from within R: RSiteSearch(Moran) and RSiteSearch(spatial autocorrelation, restrict = functions) Both seem to produce suitable results. You should also check out the spatial Task View on CRAN to get an overview of suitable packages out there that might do what you want. Choose a mirror near to you from here http://cran.r-project.org/mirrors.html and then select Task Views in menu on the left. HTH G -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Gavin Simpson [t] +44 (0)20 7679 0522 ECRC ENSIS, UCL Geography, [f] +44 (0)20 7679 0565 Pearson Building, [e] gavin.simpsonATNOSPAMucl.ac.uk Gower Street, London [w] http://www.ucl.ac.uk/~ucfagls/cv/ London, UK. WC1E 6BT. [w] http://www.ucl.ac.uk/~ucfagls/ %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] FW: Large datasets in R
Hi, I have two further comments/questions about large datasets in R. 1. Does R's ability to handle large datasets depend on the operating system's use of virtual memory? In theory, at least, VM should make the difference between installed RAM and virtual memory on a hard drive primarily a determinant of how fast R will calculate rather than whether or not it can do the calculations. However, if R has some low-level routines that have to be memory resident and use more memory as the amount of data increases, this may not hold. Can someone shed light on this? 2. Is What 64-bit versions of R are available at present? Marsh Feldman The University of Rhode Island -Original Message- From: Thomas Lumley [mailto:[EMAIL PROTECTED] Sent: Monday, July 17, 2006 3:21 PM To: Deepankar Basu Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Large datasets in R On Mon, 17 Jul 2006, Deepankar Basu wrote: Hi! I am a student of economics and currently do most of my statistical work using STATA. For various reasons (not least of which is an aversion for proprietary software), I am thinking of shifting to R. At the current juncture my concern is the following: would I be able to work on relatively large data-sets using R? For instance, I am currently working on a data-set which is about 350MB in size. Would be possible to work data-sets of such sizes using R? The answer depends on a lot of things, but most importantly 1) What you are going to do with the data 2) Whether you have a 32-bit or 64-bit version of R 3) How much memory your computer has. In a 32-bit version of R (where R will not be allowed to address more than 2-3Gb of memory) an object of size 350Mb is large enough to cause problems (see eg the R Installation and Adminstration Guide). If your 350Mb data set has lots of variables and you only use a few at a time then you may not have any trouble even on a 32-bit system once you have read in the data. If you have a 64-bit version of R and a few Gb of memory then there should be no real difficulty in working with that size of data set for most analyses. You might come across some analyses (eg some cluster analysis functions) that use n^2 memory for n observations and so break down. -thomas Thomas Lumley Assoc. Professor, Biostatistics [EMAIL PROTECTED] University of Washington, Seattle __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] RSiteSearch() not in posting guide
Hi, In a recent reply I sent to the list, my initial response was to suggest, politely that by reading the posting guide and learning to use the supplied tools one can often help oneself. I then went on to say that reading the posting guide would have led you to RSiteSearch(insert query). Luckily I went to the posting guide to check this was there before posting and much to my surprise that page does not once mention this excellent facility! In the section Do your homework before posting:, help.search() and ?foo are recommended, but these only work for installed, and possibly loaded, packages. Wouldn't it be better to suggest a user try RSiteSearch(foo) as well? This way they are far more likely to find something relevant if that something resides on CRAN or the mailing list archives. Can I suggest adding another bullet with the following text to the Do your homework before posting: section: Do RSiteSearch(keyword) with different keywords (type this at the R prompt) to search R, contributed packages and R-Help postings. To restrict the search to functions only, use RSiteSearch(keyword, restrict = functions). See ?RSiteSearch. or shorter: Do RSiteSearch(keyword) with different keywords (type this at the R prompt) to search R functions, contributed packages and R-Help postings. See ?RSiteSearch for further options and to restrict searches. G -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Gavin Simpson [t] +44 (0)20 7679 0522 ECRC ENSIS, UCL Geography, [f] +44 (0)20 7679 0565 Pearson Building, [e] gavin.simpsonATNOSPAMucl.ac.uk Gower Street, London [w] http://www.ucl.ac.uk/~ucfagls/cv/ London, UK. WC1E 6BT. [w] http://www.ucl.ac.uk/~ucfagls/ %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] RSiteSearch() not in posting guide
I think the posting guide may have been written prior to the existence of RSiteSearch as a builtin command. On 7/18/06, Gavin Simpson [EMAIL PROTECTED] wrote: Hi, In a recent reply I sent to the list, my initial response was to suggest, politely that by reading the posting guide and learning to use the supplied tools one can often help oneself. I then went on to say that reading the posting guide would have led you to RSiteSearch(insert query). Luckily I went to the posting guide to check this was there before posting and much to my surprise that page does not once mention this excellent facility! In the section Do your homework before posting:, help.search() and ?foo are recommended, but these only work for installed, and possibly loaded, packages. Wouldn't it be better to suggest a user try RSiteSearch(foo) as well? This way they are far more likely to find something relevant if that something resides on CRAN or the mailing list archives. Can I suggest adding another bullet with the following text to the Do your homework before posting: section: Do RSiteSearch(keyword) with different keywords (type this at the R prompt) to search R, contributed packages and R-Help postings. To restrict the search to functions only, use RSiteSearch(keyword, restrict = functions). See ?RSiteSearch. or shorter: Do RSiteSearch(keyword) with different keywords (type this at the R prompt) to search R functions, contributed packages and R-Help postings. See ?RSiteSearch for further options and to restrict searches. G -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Gavin Simpson [t] +44 (0)20 7679 0522 ECRC ENSIS, UCL Geography, [f] +44 (0)20 7679 0565 Pearson Building, [e] gavin.simpsonATNOSPAMucl.ac.uk Gower Street, London [w] http://www.ucl.ac.uk/~ucfagls/cv/ London, UK. WC1E 6BT. [w] http://www.ucl.ac.uk/~ucfagls/ %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] FW: Large datasets in R
In my experience, the OS's use of virtual memory is only relevant in the rough sense that the OS can store *other* running applications in virtual memory so that R can use as much of the physical memory as possible. Once R itself overflows into virtual memory it quickly becomes unusable. I'm not sure I understand your second question. As R is available in source code form, it can be compiled for many 64-bit operating systems. -roger Marshall Feldman wrote: Hi, I have two further comments/questions about large datasets in R. 1. Does R's ability to handle large datasets depend on the operating system's use of virtual memory? In theory, at least, VM should make the difference between installed RAM and virtual memory on a hard drive primarily a determinant of how fast R will calculate rather than whether or not it can do the calculations. However, if R has some low-level routines that have to be memory resident and use more memory as the amount of data increases, this may not hold. Can someone shed light on this? 2. Is What 64-bit versions of R are available at present? Marsh Feldman The University of Rhode Island -Original Message- From: Thomas Lumley [mailto:[EMAIL PROTECTED] Sent: Monday, July 17, 2006 3:21 PM To: Deepankar Basu Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Large datasets in R On Mon, 17 Jul 2006, Deepankar Basu wrote: Hi! I am a student of economics and currently do most of my statistical work using STATA. For various reasons (not least of which is an aversion for proprietary software), I am thinking of shifting to R. At the current juncture my concern is the following: would I be able to work on relatively large data-sets using R? For instance, I am currently working on a data-set which is about 350MB in size. Would be possible to work data-sets of such sizes using R? The answer depends on a lot of things, but most importantly 1) What you are going to do with the data 2) Whether you have a 32-bit or 64-bit version of R 3) How much memory your computer has. In a 32-bit version of R (where R will not be allowed to address more than 2-3Gb of memory) an object of size 350Mb is large enough to cause problems (see eg the R Installation and Adminstration Guide). If your 350Mb data set has lots of variables and you only use a few at a time then you may not have any trouble even on a 32-bit system once you have read in the data. If you have a 64-bit version of R and a few Gb of memory then there should be no real difficulty in working with that size of data set for most analyses. You might come across some analyses (eg some cluster analysis functions) that use n^2 memory for n observations and so break down. -thomas Thomas Lumley Assoc. Professor, Biostatistics [EMAIL PROTECTED] University of Washington, Seattle __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Roger D. Peng | http://www.biostat.jhsph.edu/~rpeng/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] R and DDE (Dynamic Data Exchange)
Hello Vincent, I think there is not much else than tcltk2. It is a complete implementation of DDE (client/server), so it should fit your needs. There are lots of examples on the man page. Please, if you think it could be helpful for other users, do submit examples. I think real-time acquisition of data through DDE under Windows would interest a couple of users (including myself). Best, Philippe Grosjean ..°})) ) ) ) ) ) ( ( ( ( (Prof. Philippe Grosjean ) ) ) ) ) ( ( ( ( (Numerical Ecology of Aquatic Systems ) ) ) ) ) Mons-Hainaut University, Belgium ( ( ( ( ( .. [EMAIL PROTECTED] wrote: R and DDE (Dynamic Data Exchange) Dear Rusers, I run an application (not mine) which acts as a DDE server. I would like to use R to get data from this application, say once per minute, and do some processing on it. I didn't find much info on the R DDE abilities, apart the tcltk2 package in which I will try to go deeper. I would be very thankful for any info, pointer or advice about the good ways to make R program get online data from a DDE server. Thanks Vincent __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] FW: Large datasets in R
On Tue, 18 Jul 2006, Marshall Feldman wrote: Hi, I have two further comments/questions about large datasets in R. 1. Does R's ability to handle large datasets depend on the operating system's use of virtual memory? In theory, at least, VM should make the difference between installed RAM and virtual memory on a hard drive primarily a determinant of how fast R will calculate rather than whether or not it can do the calculations. However, if R has some low-level routines that have to be memory resident and use more memory as the amount of data increases, this may not hold. Can someone shed light on this? The issue is address space, not RAM. The limits Thomas mentions are on VM, not RAM, and it is common to have at least as much RAM installed as the VM address space for a user process. There is no low-level code in R that has any idea if it is memory-resident, nor AFAIK is there any portable way to do so in a user process in a modern OS. (R is as far as possible written to C99 and POSIX standards.) 2. Is What 64-bit versions of R are available at present? Any OS with a 64-bit CPU that you can find a viable 64-bit compiler suite for. We've had 64-bit versions of R since the last millenium on Solaris, IRIX, HP-UX, OSF/1 and more recently on AIX, FreeBSD, Linux, MacOS X (on so-called G5) and probably others. The exception is probably Windows, for which there is no known free `viable 64-bit compiler suite', but it is likely that there are commercial ones. Marsh Feldman The University of Rhode Island -Original Message- From: Thomas Lumley [mailto:[EMAIL PROTECTED] Sent: Monday, July 17, 2006 3:21 PM To: Deepankar Basu Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Large datasets in R On Mon, 17 Jul 2006, Deepankar Basu wrote: Hi! I am a student of economics and currently do most of my statistical work using STATA. For various reasons (not least of which is an aversion for proprietary software), I am thinking of shifting to R. At the current juncture my concern is the following: would I be able to work on relatively large data-sets using R? For instance, I am currently working on a data-set which is about 350MB in size. Would be possible to work data-sets of such sizes using R? The answer depends on a lot of things, but most importantly 1) What you are going to do with the data 2) Whether you have a 32-bit or 64-bit version of R 3) How much memory your computer has. In a 32-bit version of R (where R will not be allowed to address more than 2-3Gb of memory) an object of size 350Mb is large enough to cause problems (see eg the R Installation and Adminstration Guide). If your 350Mb data set has lots of variables and you only use a few at a time then you may not have any trouble even on a 32-bit system once you have read in the data. If you have a 64-bit version of R and a few Gb of memory then there should be no real difficulty in working with that size of data set for most analyses. You might come across some analyses (eg some cluster analysis functions) that use n^2 memory for n observations and so break down. -thomas Thomas Lumley Assoc. Professor, Biostatistics [EMAIL PROTECTED] University of Washington, Seattle __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] using split.screen?
Hello. I am having trouble understanding the use of split.screen. I want to divide the device surface first into 4 equal screens: split.screen(figs=c(2,2)) This works. I next want to subdivide each of these 4 screens into 10 subscreens. I do, for the first of these 4 screens: screen(1,new=T) and then: split.screen(figs=c(10,2)) My understanding is that this should split screen 1 (i.e. top right) into 10 screens within its area. However, this does not work and I get the following error message: Error in plot.new() : figure margins too large Does anyone know what I am doing wrong? Bill Shipley [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Inflated Array
Hi R-users! I am trying to create a what I call inflated array (maybe there is already some other name for that). It is an array that changes dinamically its dimensions, e.g. the higher the number of third dimensions, the more rows in the array. So for example the array could look like the following: , ,1 1 2 3 4 , , 2 5 6 7 8 9 10 11 12 , , 3 13 14 15 16 17 18 19 20 21 22 23 24 Has anybody a comment on how to build this in the most efficient manner (I know i could do looping)? Thanks! Hadassa __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] R and DDE (Dynamic Data Exchange)
You can access DDE via COM as in this example which uses DDE to open an Excel file. Note that Excel also supports COM directly and normally one would use COM with Excel, not DDE, so you might check if your application also supports COM. # opens an excel spreadsheet c:\test.xls using dde library(RDCOMClient) sh - COMCreate(Shell.Application) sh$Namespace(C:\\)$ParseName(test.xls)$InvokeVerb(Open) Also if you are going to access DDE via COM or just COM also check out the rcom package which is similar to RDCOMClient. On 7/17/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: R and DDE (Dynamic Data Exchange) Dear Rusers, I run an application (not mine) which acts as a DDE server. I would like to use R to get data from this application, say once per minute, and do some processing on it. I didn't find much info on the R DDE abilities, apart the tcltk2 package in which I will try to go deeper. I would be very thankful for any info, pointer or advice about the good ways to make R program get online data from a DDE server. Thanks Vincent __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] using split.screen?
On Tue, 18 Jul 2006, Bill Shipley wrote: Hello. I am having trouble understanding the use of split.screen. I want to divide the device surface first into 4 equal screens: split.screen(figs=c(2,2)) This works. I next want to subdivide each of these 4 screens into 10 subscreens. I do, for the first of these 4 screens: screen(1,new=T) and then: split.screen(figs=c(10,2)) My understanding is that this should split screen 1 (i.e. top right) That is screen 2: 1 is top left. into 10 screens within its area. However, this does not work and I get the following error message: Error in plot.new() : figure margins too large well, only when you try to plot on one of those screens. Does anyone know what I am doing wrong? I suspect not using a big enough device region to allow such small plot regions. You are asking for 20 rows in your device region. Oh, and sending HTML mail. Bill Shipley [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html ^ and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] using split.screen? [Broadcast]
This means that the margins for 10 screens would take up more room than you have - essentially the plot area is being squeezed to nothing. You can try reducing your margins using par. Also, it looks like you're trying to split into 20 screens there. Hope this helps, Matt -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Bill Shipley Sent: Tuesday, July 18, 2006 10:02 AM To: R help list Subject: [R] using split.screen? [Broadcast] Hello. I am having trouble understanding the use of split.screen. I want to divide the device surface first into 4 equal screens: split.screen(figs=c(2,2)) This works. I next want to subdivide each of these 4 screens into 10 subscreens. I do, for the first of these 4 screens: screen(1,new=T) and then: split.screen(figs=c(10,2)) My understanding is that this should split screen 1 (i.e. top right) into 10 screens within its area. However, this does not work and I get the following error message: Error in plot.new() : figure margins too large Does anyone know what I am doing wrong? Bill Shipley [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] FW: Large datasets in R
Hi, I have a related question. How differently do other statistical softwares handle large data? The original post claims that 350 MB is fine on Stata. Some one suggested S-Plus. I have heard people say that SAS can handle large data sets. Why can others do it and R seem to have a problem? Don't these softwares load the data onto RAM. -- Ritwik Sinha Graduate Student Epidemiology and Biostatistics Case Western Reserve University http://darwin.cwru.edu/~rsinha __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Nested functions
On Mon, 17 Jul 2006, John Wiedenhoeft wrote:
Hi there,
I'm having myself a hard time writing an algorithm for finding patterns
within a given melody. In a vector I'd like to find ALL sequences that
occur at least twice, without having to check all possible patterns via
pattern matching.
Another approach, which works for not-too-long vectors like you have is:
n - length(v)
matches - outer(v, v, ==) outer(1:n,1:n,=)
Now matches has TRUE where v[i]==v[j]. For a longer match you would also
need v[i+1]==v[j+1] and so on, making a diagonal line through the matrix.
Diagonal lines are hard, so let's turn them into horizontal lines
matches - matrix(cbind(matches, FALSE), ncol=n)
now row i+1 column j of matches is TRUE for a single entry match starting
at position j at a separation of i. If there is a match of length 2, then
column j+1 will also be TRUE, and so on.
Now rle() applied to a row will return the lengths of consecutive
sequences of TRUE and FALSE. The lengths of consecutive sequences of TRUE
are the lengths of the matches. To get rid of trivial matches of length
less than 2 do
match2 - t(apply(matches,1,function(row){
r-rle(row)
r$values[r$lengths2]-FALSE
inverse.rle(r)
}))
And finally, to extract the matches
results - apply(match2, 1, function(row){
r-rle(row)
n-length(r$lengths)
ends-cumsum(r$lengths)
starts-cumsum(c(1,r$lengths))[1:n]
list(starts[r$values],ends[r$values])
})
for starts and ends of matches or
results - apply(match2, 1, function(row){
r-rle(row)
n-length(r$lengths)
ends-cumsum(r$lengths)[r$values]
starts-cumsum(c(1,r$lengths))[1:n][r$values]
mapply(function(stt,end) v[stt:end],starts,ends,
SIMPLIFY=FALSE)
})
to get a list of the actual matching sequences.
-thomas
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] FW: Large datasets in R
S-Plus stores objects as files whereas R stores them in memory. SAS was developed many years ago when optimizing computer resources was more important than it is now. On 7/18/06, Ritwik Sinha [EMAIL PROTECTED] wrote: Hi, I have a related question. How differently do other statistical softwares handle large data? The original post claims that 350 MB is fine on Stata. Some one suggested S-Plus. I have heard people say that SAS can handle large data sets. Why can others do it and R seem to have a problem? Don't these softwares load the data onto RAM. -- Ritwik Sinha Graduate Student Epidemiology and Biostatistics Case Western Reserve University http://darwin.cwru.edu/~rsinha __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] RSiteSearch() not in posting guide
Gabor == Gabor Grothendieck [EMAIL PROTECTED] on Tue, 18 Jul 2006 09:29:07 -0400 writes: Gabor I think the posting guide may have been written prior Gabor to the existence of RSiteSearch as a builtin command. yes, that's correct. Thank you, Gavin, for the suggestion. I have added such a bullet. Visible now at svn.r-project.org, and from tomorrow at the URL at the tail of every R-help message. Martin Gabor On 7/18/06, Gavin Simpson [EMAIL PROTECTED] Gabor wrote: Hi, In a recent reply I sent to the list, my initial response was to suggest, politely that by reading the posting guide and learning to use the supplied tools one can often help oneself. I then went on to say that reading the posting guide would have led you to RSiteSearch(insert query). Luckily I went to the posting guide to check this was there before posting and much to my surprise that page does not once mention this excellent facility! In the section Do your homework before posting:, help.search() and ?foo are recommended, but these only work for installed, and possibly loaded, packages. Wouldn't it be better to suggest a user try RSiteSearch(foo) as well? This way they are far more likely to find something relevant if that something resides on CRAN or the mailing list archives. Can I suggest adding another bullet with the following text to the Do your homework before posting: section: Do RSiteSearch(keyword) with different keywords (type this at the R prompt) to search R, contributed packages and R-Help postings. To restrict the search to functions only, use RSiteSearch(keyword, restrict = functions). See ?RSiteSearch. or shorter: Do RSiteSearch(keyword) with different keywords (type this at the R prompt) to search R functions, contributed packages and R-Help postings. See ?RSiteSearch for further options and to restrict searches. G -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Gavin Simpson [t] +44 (0)20 7679 0522 ECRC ENSIS, UCL Geography, [f] +44 (0)20 7679 0565 Pearson Building, [e] gavin.simpsonATNOSPAMucl.ac.uk Gower Street, London [w] http://www.ucl.ac.uk/~ucfagls/cv/ London, UK. WC1E 6BT. [w] http://www.ucl.ac.uk/~ucfagls/ %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] A contingency table of counts by case
Here is an example of the data.frame that I have, df-data.frame(case=rep(1:5,each=9),id=rep(1:9,times=5),x=round(runif(length(rep(1:5,each=9) case represents the cases, id the persons, and x is the binary state. I would like to know in how many cases any two persons a. both have 1, b. the first has 0 - the second has 1, c. the first has 0 - the second has 0, d. both have 0. There will be choose(9,2) sums, denoted s_ij for 1=ij=9, where i and j are running indices for an id-pair. Please help, this is way beyond my knowledge of R! Thank you, Serguei Kaniovski -- ___ Austrian Institute of Economic Research (WIFO) Name: Serguei Kaniovski P.O.Box 91 Tel.: +43-1-7982601-231 Arsenal Objekt 20 Fax: +43-1-7989386 1103 Vienna, Austria Mail: [EMAIL PROTECTED] http://www.wifo.ac.at/Serguei.Kaniovski __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] FW: Large datasets in R
On Tue, 18 Jul 2006, Ritwik Sinha wrote: Hi, I have a related question. How differently do other statistical softwares handle large data? The original post claims that 350 MB is fine on Stata. Some one suggested S-Plus. I have heard people say that SAS can handle large data sets. Why can others do it and R seem to have a problem? Don't these softwares load the data onto RAM. Stata does load the data into RAM and does have limits for the same reason that R does. However, Stata has a less flexible representation of its data (basically one rectangular dataset) and so it can handle somewhat larger data sets for any given memory size. For example, even with 512Gb of memory a 350Mb data set might be usable in Stata and with 1Gb it would certainly be. Stata is also faster for a given memory load, apparently because of its simpler language design [some evidence for this is that the recent language additions to support flexible graphics run rather more slowly than eg lattice in R]. The other approach is to write the estimation routines so that only part of the data need be in memory at a given time. *Some* procedures in SAS and SPSS work this way, and this is the idea of the S-PLUS 7.0 system for handling large data sets. This approach requires the programmer to handle the reading of sections of code from memory, something that can only be automated to a limited extent. People have used R in this way, storing data in a database and reading it as required. There are also some efforts to provide facilities to support this sort of programming (such as the current project funded by Google Summer of Code: http://tolstoy.newcastle.edu.au/R/devel/06/05/5525.html). One reason there isn't more of this is that relying on Moore's Law has worked very well over the years. -thomas Thomas Lumley Assoc. Professor, Biostatistics [EMAIL PROTECTED] University of Washington, Seattle __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Nested functions
Am Dienstag, den 18.07.2006, 22:09 -0400 schrieb Jim Lemon:
Hi John,
Minor bug - I zeroed the hit counter in the wrong place.
find.replay-function(tunestring,maxlen) {
return(matchlist)
}
Dear Jim,
many, many thanks for your effords :-D!!! Your program is great and very
elegant (guess I was thinking to mathematical and iterative...). I have
made some minor adjustments to it. Most important, I have changed
starttest to startpat+1, since patterns can overlap of course (
consists of 2 times 111). Hope this doesn't cause any inconsistency.
Here is what I'm using now:
CODECODECODECODECODECODECODECODECODECODECODECODECODECODECODE
find.replay-function(tunestring,filename,maxlen)
{
tunelen-length(tunestring)
if(missing(maxlen)) maxlen-floor(tunelen/2)
if(missing(filename)) filename-output
matchlist - list()
startpat-1
endpat-2
finishpos-tunelen-maxlen
pattern-tunestring[startpat:endpat]
patlen-length(pattern)
while(patlen = maxlen)
{
while(endpat tunelen-patlen)
{
starttest-startpat+1 #changed from endpat+1 to detect
overlapping
patterns
endtest-starttest+(endpat-startpat)
# step through the rest of tunestring with this pattern
while(endtest = tunelen)
{
testpat-tunestring[starttest:endtest]
if(identical(pattern,testpat))
{
m - 0;
w - 0;
for(k in 1:patlen)
{m - pattern[k]*10^(patlen-k)+m;}
for(l in 1:patlen)
{w - testpat[l]*10^(patlen-l)+w;}
#just in case... ;-)
if (m!=w)
{
warn - paste(Unmatching patterns, m,
and, w, detected
errorneously!)
print(warn)
write.table(warn, file=filename,
append=TRUE, sep=\t, eol=\n,
row.names=FALSE, col.names=FALSE);
}
p - c(patlen, m, startpat, starttest);
write.table(t(p), file=filename, append=TRUE,
sep=\t, eol=\n,
row.names=FALSE, col.names=FALSE);
# print(p);
}
# step to the next substring
starttest-starttest+1
endtest-endtest+1
}
# now move pattern one step along the string
startpat-startpat+1
endpat-endpat+1
pattern-tunestring[startpat:endpat]
}
# now increase the length of the pattern
startpat- 1
endpat-startpat+patlen
pattern-tunestring[startpat:endpat]
patlen-length(pattern)
}
print(paste(Output written to file ', filename, '. Have fun!,
sep=))
}
ENDENDENDENDENDENDENDENDENDENDENDENDENDENDENDENDENDENDENDEND
Again, many, many thanks. You brightened up my day!
Cheers,
John
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] A contingency table of counts by case
library(gtools) apply(combinations(9,2), 1, function(x) with(df[df$id %in% x, ], table(x, id))) --- Jacques VESLOT CNRS UMR 8090 I.B.L (2ème étage) 1 rue du Professeur Calmette B.P. 245 59019 Lille Cedex Tel : 33 (0)3.20.87.10.44 Fax : 33 (0)3.20.87.10.31 http://www-good.ibl.fr --- Serguei Kaniovski a écrit : Here is an example of the data.frame that I have, df-data.frame(case=rep(1:5,each=9),id=rep(1:9,times=5),x=round(runif(length(rep(1:5,each=9) case represents the cases, id the persons, and x is the binary state. I would like to know in how many cases any two persons a. both have 1, b. the first has 0 - the second has 1, c. the first has 0 - the second has 0, d. both have 0. There will be choose(9,2) sums, denoted s_ij for 1=ij=9, where i and j are running indices for an id-pair. Please help, this is way beyond my knowledge of R! Thank you, Serguei Kaniovski __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Survey-weighted ordered logistic regression
Hi, I am trying to fit a model with an ordered response variable (3 levels) and 13 predictor variables. The sample has complex survey design and I've used 'svydesign' command from the survey package to specify the sampling design. After reading the manual of 'svyglm' command, I've found that you can fit a logistic regression (binary response variable) by specifying the family=binomial in svyglm function. However I'm unable to fit an ordered logistic model in 'svyglm function'. Any help with this will be appreciated. Thanks, Debarchana. -- Debarchana Ghosh Research Assistant, Department of Geography University of Minnesota. PH: 8143607580 www.tc.umn.edu/~ghos0033/ [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] I think this is a bug
Hello!
I work with:
R : Copyright 2006, The R Foundation for
Statistical Computing
Version 2.3.1 (2006-06-01)
On Windows XP Professional (Version 2002) SP2
I think there is a bug in the conditional
execution if (expr1) {expr2} else {expr3}
If I try:
if (expr1) expr2 else expr3
it works well but when I put the expression expr2
and expr3 between {} I receive an error message
like this one:
Erreur dans parse(file, n = -1, NULL, ?) :
erreur de syntaxe à la ligne
4: }
5: else
...which translated in english gives:
Error in parse(file, n = -1, NULL, ?) : syntax
error at the line
4: }
5: else
Maybe, there is something I don't understand. I
should be very grateful if you would help me to
solve this issue.
Best regards,
Xavier
___
Xavier Barron
20, rue de la Pierre Levée
75011 Paris
0143381141 / 0675062109
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] How can I extract information from list which class is nls
Hello! I work with : R : Copyright 2006, The R Foundation for Statistical Computing Version 2.3.1 (2006-06-01) On Windows XP Professional (Version 2002) SP2. At this moment I use the function nls combined with a selfStar model (SSmicmen, related to Michaelis-Menten equation, and provided by the stats package). When I realise the following operation (cf. p 59 of the An Introduction to R manual, http://www.r-project.org/, for more details): fit-nls(y~SSmicmen(x, Vm, K), df) summary(fit) I obtain the values of Vm and K. The object fit is a list of the class nls. However I cannot identify the objects which are inside of fit and which contain the values of Vm and K. Actually, I would like to extract these values to introduce them into new objects but I don't know how?! Maybe, somebody could help me to solve this problem. It would be very helpful for me. Best regards, Xavier __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Inflated Array
Hadassa Brunschwig dassybr at gmail.com writes: Hi R-users! I am trying to create a what I call inflated array (maybe there is already some other name for that). It is an array that changes dinamically its dimensions, e.g. the higher the number of third dimensions, the more rows in the array. base R doesn't have a data structure for this: arrays in R must not be ragged (i.e., every sub-array must have the same dimensions). So you would need to use a list of matrices in your example. I don't quite know the logic that you're using to decide what goes in each sub-table. The basic example you give could be built by hand as follows: list(1:4,matrix(5:12,byrow=TRUE,ncol=4), matrix(13:24,byrow=TRUE,ncol=4)) if you had used scan() or something else to get a long, flat vector (x) and also had a vector (v) that indicated the number of rows in each sub-table: x - 1:24 nrow - c(1,2,3) ncol - 4 ind - rep(1:length(nrow),ncol*nrow) lapply(split(x,ind), matrix,ncol=ncol,byrow=TRUE) seems to work. cheers Ben Bolker __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] R and DDE (Dynamic Data Exchange)
Thanks Gabor and Philippe. Special thanks to Philippe for his tcltk2 nice job. I'm testing differents approaches for my problem. I'll return info if I use DDE + tcltk2. Vincent __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Endogenous Tobit/Probit model
Hi I would like to know if there is a package that allows me to implement endogenous Tobit/Probit model. Example: res1 - rnorm(N); res2 - res1*0.5 + rnorm(N) x - z[,1]*2 + res1; ys - x*b + res2; d - (ys0); #dummy variable y - d*ys; y is censored and x is correlated with res1. z is my instrument. (continuously observed) . Therefore survreg(Surv(y, y 0, type ='left')~ 0 + x , dist = 'gaussian') will give biased estimates. I also need to know the variance-covariance matrix for the parameters of the model Thanks for any information Leandro __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] I think this is a bug
On Tue, 18 Jul 2006, Xavier Barron wrote:
Hello!
I work with:
R : Copyright 2006, The R Foundation for
Statistical Computing
Version 2.3.1 (2006-06-01)
On Windows XP Professional (Version 2002) SP2
I think there is a bug in the conditional
execution if (expr1) {expr2} else {expr3}
If I try:
if (expr1) expr2 else expr3
it works well but when I put the expression expr2
and expr3 between {} I receive an error message
like this one:
It's not a bug. You need the } on the same line as the else
if (expr1){
expr2
} else {
expr3
}
as otherwise R doesn't know there is going to be an 'else'.
-thomas
Erreur dans parse(file, n = -1, NULL, ?) :
erreur de syntaxe à la ligne
4: }
5: else
...which translated in english gives:
Error in parse(file, n = -1, NULL, ?) : syntax
error at the line
4: }
5: else
Maybe, there is something I don't understand. I
should be very grateful if you would help me to
solve this issue.
Best regards,
Xavier
___
Xavier Barron
20, rue de la Pierre Levée
75011 Paris
0143381141 / 0675062109
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Thomas Lumley Assoc. Professor, Biostatistics
[EMAIL PROTECTED] University of Washington, Seattle__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] How can I extract information from list which class is nls
Hi your fit is an object (list) and you could use some functions like summary or coef to extract usefull information from it or you can call its components on your own. DNase1 - subset(DNase, Run == 1) ## using a selfStart model fm1DNase1 - nls(density ~ SSlogis(log(conc), Asym, xmid, scal), DNase1) summary(fm1DNase1) Formula: density ~ SSlogis(log(conc), Asym, xmid, scal) Parameters: Estimate Std. Error t value Pr(|t|) Asym 2.345180.07815 30.01 2.17e-13 *** xmid 1.483090.08135 18.23 1.22e-10 *** scal 1.041460.03227 32.27 8.51e-14 *** --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 Residual standard error: 0.01919 on 13 degrees of freedom coef(fm1DNase1) Asym xmid scal 2.345180 1.483090 1.041455 coef(fm1DNase1)[1] Asym 2.34518 HTH Petr On 18 Jul 2006 at 18:06, Xavier Barron wrote: Date sent: Tue, 18 Jul 2006 18:06:00 +0200 (CEST) From: Xavier Barron [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Subject:[R] How can I extract information from list which class is nls Hello! I work with : R : Copyright 2006, The R Foundation for Statistical Computing Version 2.3.1 (2006-06-01) On Windows XP Professional (Version 2002) SP2. At this moment I use the function nls combined with a selfStar model (SSmicmen, related to Michaelis-Menten equation, and provided by the stats package). When I realise the following operation (cf. p 59 of the An Introduction to R manual, http://www.r-project.org/, for more details): fit-nls(y~SSmicmen(x, Vm, K), df) summary(fit) I obtain the values of Vm and K. The object fit is a list of the class nls. However I cannot identify the objects which are inside of fit and which contain the values of Vm and K. Actually, I would like to extract these values to introduce them into new objects but I don't know how?! Maybe, somebody could help me to solve this problem. It would be very helpful for me. Best regards, Xavier __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Petr Pikal [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] I think this is a bug
Hi
not at all
works for me
a-1
b-2
if (ab) {print(Hallo)} else {print(OK)}
[1] OK
if (ab) {print(Hallo)} else {print(OK)}
[1] Hallo
you probably started your else on new line e.g.
if (ab) {print(Hallo)} {
Error: syntax error in if (ab) {print(Hallo)} {
if (ab)
+ {print(Hallo)} else
+ {print(OK)}
[1] Hallo
HTH
Petr
On 18 Jul 2006 at 17:43, Xavier Barron wrote:
Date sent: Tue, 18 Jul 2006 17:43:42 +0200 (CEST)
From: Xavier Barron [EMAIL PROTECTED]
To: r-help@stat.math.ethz.ch
Subject:[R] I think this is a bug
Hello!
I work with:
R : Copyright 2006, The R Foundation for
Statistical Computing
Version 2.3.1 (2006-06-01)
On Windows XP Professional (Version 2002) SP2
I think there is a bug in the conditional
execution if (expr1) {expr2} else {expr3}
If I try:
if (expr1) expr2 else expr3
it works well but when I put the expression expr2
and expr3 between {} I receive an error message
like this one:
Erreur dans parse(file, n = -1, NULL, ?) :
erreur de syntaxe ŕ la ligne
4: }
5: else
...which translated in english gives:
Error in parse(file, n = -1, NULL, ?) : syntax
error at the line
4: }
5: else
Maybe, there is something I don't understand. I
should be very grateful if you would help me to
solve this issue.
Best regards,
Xavier
___
Xavier Barron
20, rue de la Pierre Levée
75011 Paris
0143381141 / 0675062109
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html and provide commented,
minimal, self-contained, reproducible code.
Petr Pikal
[EMAIL PROTECTED]
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Output and Word
Others have suggested using R2HTML (which is a good option). Another option is to use Sweave and specifically the new odfWeave package for R. This works on OpenOffice files rather than word files (but OpenOffice http://www.openoffice.org/ can inport and export word documents). The basic idea is to write your report in OpenOffice (or LaTeX or HTML), but anywhere that you want statisticial output (graphs, tables) you include instead the R code to produce the table, graph, or whatever. Run this file through R (using Sweave or odfWeave) and the resulting file has replaced all the code segments with their output. The documentation with odfWeave has examples. Hope this helps, -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare [EMAIL PROTECTED] (801) 408-8111 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of sharon snowdon Sent: Monday, July 17, 2006 2:36 PM To: r-help@stat.math.ethz.ch Subject: [R] Output and Word Hi I have just started to have a look at R. I have used most stats software packages and can use perl, visual basic etc. I am interested in how well it handles lots of output e.g. tables or charts. How would you get lots of output most easily and quickly into a Word document? Sharon Snowdon - -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Sweave and multipage lattice
Dear R-Listeners, as the Sweave faq says: http://www.ci.tuwien.ac.at/~leisch/Sweave/FAQ.html creating several figures from one figure chunk does not work, and for standard graphics, a workaround is given. Now I have a multipage trellis plot with an a-priori unknown number of pages, and I don't see an elegant way of dividing it up into multiple pdf-files. I noted there is a page event handler in the ... parameters, which would provide a handle to open/close the file. Any good idea would be appreciated. Dieter __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Survey-weighted ordered logistic regression
On Tue, 18 Jul 2006, Debarchana Ghosh wrote: Hi, I am trying to fit a model with an ordered response variable (3 levels) and 13 predictor variables. The sample has complex survey design and I've used 'svydesign' command from the survey package to specify the sampling design. After reading the manual of 'svyglm' command, I've found that you can fit a logistic regression (binary response variable) by specifying the family=binomial in svyglm function. However I'm unable to fit an ordered logistic model in 'svyglm function'. You can do this most easily using withReplicates(). If you have a survey object created with svydesign, use as.svrepdesign() to turn it into a replicate-weights object and then do eg library(MASS) ologit - withReplicates( repwtdesign, quote(coef(polr(y~x1+x2+x3+x4, weights=.weights You will get a lot of (harmless) warnings about non-integer #successes in a binomial glm!. This will not give you standard errors for the intercept terms - if you want them you can do ologit.int - withReplicates( repwtdesign, quote(polr(y~x1+x2+x3+x4, weights=.weights)$zeta)) You could also use svymle() and get Taylor linearisation standard errors, but that would require writing out the loglikelihood and its gradient, which is tedious (although you could borrow most of it from MASS::polr) -thomas Thomas Lumley Assoc. Professor, Biostatistics [EMAIL PROTECTED] University of Washington, Seattle __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Reconfiguring wide frame to long frame
Greetings, fellow R'ers. How can I get this frame in R: ID meas ID.1 meas.1 1 1.13 1.2 2 2.14 2.2 to look like this (stacking): ID meas 1 1.1 2 2.1 3 1.2 4 2.2 It's not really the reshape function (or is it?) because we can consider the additional columns, viz., ID.1 and meas.1, as independent of ID and meas so it is basically a stacking problem (no longitudinal component). I can't seem to find a good example to do this in the docs. Thanks for your help. Regards, Jesse Jesse A. Canchola Biostatistician III Bayer Healthcare 725 Potter St. Berkeley, CA 94710 P: 510.705.5855 F: 510.705.5718 E: [EMAIL PROTECTED] ___ The information contained in this e-mail is for the exclusive use of the intended recipient(s) and may be confidential, proprietary, and/or legally privileged. Inadvertent disclosure of this message does not constitute a waiver of any privilege. If you receive this message in error, please do not directly or indirectly use, print, copy, forward, or disclose any part of this message. Please also delete this e-mail and all copies and notify the sender. Thank you. For alternate languages please go to http://bayerdisclaimer.bayerweb.com ___ [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Large datasets in R
Thanks a lot for all the responses. The general drift of all the messages was the suggestion to use some database management package that has a nice interface with R; and most of the suggestions pointed in the direction of SQL. I will look into the SQL package and start learning to use it along with R. Thanks once again for all your suggestions. Deepankar __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Reconfiguring wide frame to long frame
Try this: # set up test data Lines - ID meas ID.1 meas.1 1 1.13 1.2 2 2.14 2.2 DF - read.table(textConnection(Lines), header = TRUE) # reshape matrix(t(DF), nc = 2, byrow = TRUE, dimnames = list(NULL, colnames(DF)[1:2])) On 7/18/06, Jesse Albert Canchola [EMAIL PROTECTED] wrote: Greetings, fellow R'ers. How can I get this frame in R: ID meas ID.1 meas.1 1 1.13 1.2 2 2.14 2.2 to look like this (stacking): ID meas 1 1.1 2 2.1 3 1.2 4 2.2 It's not really the reshape function (or is it?) because we can consider the additional columns, viz., ID.1 and meas.1, as independent of ID and meas so it is basically a stacking problem (no longitudinal component). I can't seem to find a good example to do this in the docs. Thanks for your help. Regards, Jesse Jesse A. Canchola Biostatistician III Bayer Healthcare 725 Potter St. Berkeley, CA 94710 P: 510.705.5855 F: 510.705.5718 E: [EMAIL PROTECTED] ___ The information contained in this e-mail is for the exclusive use of the intended recipient(s) and may be confidential, proprietary, and/or legally privileged. Inadvertent disclosure of this message does not constitute a waiver of any privilege. If you receive this message in error, please do not directly or indirectly use, print, copy, forward, or disclose any part of this message. Please also delete this e-mail and all copies and notify the sender. Thank you. For alternate languages please go to http://bayerdisclaimer.bayerweb.com ___ [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] How best to deal with returned errors?
Hi, What is the best general strategy to prevent returned errors from interrupting whatever it is I am running? Only by using options()? This is a problem for me in 2 particular cases: (i) An error breaking my loops. (ii) I would like to run some regressions where it automatically will try a specified number of increasingly robust options until the regression doesnt fail. Using try(), especially for the second case, seems a bit clunky for my needs. As Ive never really automated the handling of my errors I was wondering if there is perhaps a bit of simple wisdom or pointers in the right direction on the matter. Thanks, Quin -- Checked by AVG Free Edition. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Reconfiguring wide frame to long frame
Sorry, in looking at this again my previous code did not give the same ordering you indicated. Instead using the same DF try this: rbind(as.matrix(DF[,1:2]), as.matrix(DF[,3:4])) Both this and the last piece of code produce matrices so use as.data.frame if you want a data frame. On 7/18/06, Gabor Grothendieck [EMAIL PROTECTED] wrote: Try this: # set up test data Lines - ID meas ID.1 meas.1 1 1.13 1.2 2 2.14 2.2 DF - read.table(textConnection(Lines), header = TRUE) # reshape matrix(t(DF), nc = 2, byrow = TRUE, dimnames = list(NULL, colnames(DF)[1:2])) On 7/18/06, Jesse Albert Canchola [EMAIL PROTECTED] wrote: Greetings, fellow R'ers. How can I get this frame in R: ID meas ID.1 meas.1 1 1.13 1.2 2 2.14 2.2 to look like this (stacking): ID meas 1 1.1 2 2.1 3 1.2 4 2.2 It's not really the reshape function (or is it?) because we can consider the additional columns, viz., ID.1 and meas.1, as independent of ID and meas so it is basically a stacking problem (no longitudinal component). I can't seem to find a good example to do this in the docs. Thanks for your help. Regards, Jesse Jesse A. Canchola Biostatistician III Bayer Healthcare 725 Potter St. Berkeley, CA 94710 P: 510.705.5855 F: 510.705.5718 E: [EMAIL PROTECTED] ___ The information contained in this e-mail is for the exclusive use of the intended recipient(s) and may be confidential, proprietary, and/or legally privileged. Inadvertent disclosure of this message does not constitute a waiver of any privilege. If you receive this message in error, please do not directly or indirectly use, print, copy, forward, or disclose any part of this message. Please also delete this e-mail and all copies and notify the sender. Thank you. For alternate languages please go to http://bayerdisclaimer.bayerweb.com ___ [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] R and DDE (Dynamic Data Exchange)
I am thrilled to learn tcltk2 has DDE capability. It is the piece I have been needing to make ESS work directly with the RGUI on Windows. GNU emacs on Windows has a ddeclient, but no access to COM. So if R, or tcltk2 talking in both directions to R, has a ddeserver, all should be possible. I will be reading the documentation closely in a few weeks to tie it together and then intend to make it happen. Do you, or any other list member, have a sense of the size, complexity, ease, magnitude of the task I just defined? Any advice as I get started on it? Rich __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] bilinear regression
I think this is an easy question, but I would be grateful for any advice on how to implement this in R. I simply have a response variable (y) that I am trying to predict with one explanatory variable (x) but the shape of the scatter plot is distinctly bilinear. It would be best described by two straight lines. Is there a way of fitting a linear model to give me a bilinear fit and (more importantly) automatically determine the 'cut off' point? I would also want some statistic to convince myself that the bilinear fit is better. Many thanks for your help. David Dr. David Crabb Department of Optometry and Visual Science, City University, Northampton Square, London EC1V OHB Tel: 44 207 040 0191 [EMAIL PROTECTED] http://www.city.ac.uk/optometry/html/david_crabb.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Reconfiguring wide frame to long frame
Many thanks, Gabor. That worked great! I'm ecstatic. Best regards, Jesse Gabor Grothendieck [EMAIL PROTECTED] 07/18/2006 10:12 AM To Jesse Albert Canchola [EMAIL PROTECTED] cc r-help@stat.math.ethz.ch Subject Re: [R] Reconfiguring wide frame to long frame Try this: # set up test data Lines - ID meas ID.1 meas.1 1 1.13 1.2 2 2.14 2.2 DF - read.table(textConnection(Lines), header = TRUE) # reshape matrix(t(DF), nc = 2, byrow = TRUE, dimnames = list(NULL, colnames(DF)[1:2])) On 7/18/06, Jesse Albert Canchola [EMAIL PROTECTED] wrote: Greetings, fellow R'ers. How can I get this frame in R: ID meas ID.1 meas.1 1 1.13 1.2 2 2.14 2.2 to look like this (stacking): ID meas 1 1.1 2 2.1 3 1.2 4 2.2 It's not really the reshape function (or is it?) because we can consider the additional columns, viz., ID.1 and meas.1, as independent of ID and meas so it is basically a stacking problem (no longitudinal component). I can't seem to find a good example to do this in the docs. Thanks for your help. Regards, Jesse Jesse A. Canchola Biostatistician III Bayer Healthcare 725 Potter St. Berkeley, CA 94710 P: 510.705.5855 F: 510.705.5718 E: [EMAIL PROTECTED] ___ The information contained in this e-mail is for the exclusive use of the intended recipient(s) and may be confidential, proprietary, and/or legally privileged. Inadvertent disclosure of this message does not constitute a waiver of any privilege. If you receive this message in error, please do not directly or indirectly use, print, copy, forward, or disclose any part of this message. Please also delete this e-mail and all copies and notify the sender. Thank you. For alternate languages please go to http://bayerdisclaimer.bayerweb.com ___ [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Reconfiguring wide frame to long frame
Thanks, Gabor. Since the data stacking components are independent, that didn't matter much but I am grateful for your follow-up code to match the desired output specifically. Regards, Jesse Gabor Grothendieck [EMAIL PROTECTED] 07/18/2006 10:36 AM To Jesse Albert Canchola [EMAIL PROTECTED] cc r-help@stat.math.ethz.ch Subject Re: [R] Reconfiguring wide frame to long frame Sorry, in looking at this again my previous code did not give the same ordering you indicated. Instead using the same DF try this: rbind(as.matrix(DF[,1:2]), as.matrix(DF[,3:4])) Both this and the last piece of code produce matrices so use as.data.frame if you want a data frame. On 7/18/06, Gabor Grothendieck [EMAIL PROTECTED] wrote: Try this: # set up test data Lines - ID meas ID.1 meas.1 1 1.13 1.2 2 2.14 2.2 DF - read.table(textConnection(Lines), header = TRUE) # reshape matrix(t(DF), nc = 2, byrow = TRUE, dimnames = list(NULL, colnames(DF)[1:2])) On 7/18/06, Jesse Albert Canchola [EMAIL PROTECTED] wrote: Greetings, fellow R'ers. How can I get this frame in R: ID meas ID.1 meas.1 1 1.13 1.2 2 2.14 2.2 to look like this (stacking): ID meas 1 1.1 2 2.1 3 1.2 4 2.2 It's not really the reshape function (or is it?) because we can consider the additional columns, viz., ID.1 and meas.1, as independent of ID and meas so it is basically a stacking problem (no longitudinal component). I can't seem to find a good example to do this in the docs. Thanks for your help. Regards, Jesse Jesse A. Canchola Biostatistician III Bayer Healthcare 725 Potter St. Berkeley, CA 94710 P: 510.705.5855 F: 510.705.5718 E: [EMAIL PROTECTED] ___ The information contained in this e-mail is for the exclusive use of the intended recipient(s) and may be confidential, proprietary, and/or legally privileged. Inadvertent disclosure of this message does not constitute a waiver of any privilege. If you receive this message in error, please do not directly or indirectly use, print, copy, forward, or disclose any part of this message. Please also delete this e-mail and all copies and notify the sender. Thank you. For alternate languages please go to http://bayerdisclaimer.bayerweb.com ___ [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] bilinear regression
It appears that you might have a latent (hidden) explanatory variable that causes the two-population appearance. If you have some ideas on what that other factor might be, you could try two separate linear regressions for each value of the latent factor and compare the slopes and intercepts. You can then do some formal tests on the slopes and intercepts to see if you can further simplify the model. Depending on what you find, you can formulate a linear regression model that incorporates such dependence on the slopes or intercepts to fit the bilinear trend. You might find helpful the discussion and example in Ch.10 of Venables Ripley, 4th ed, that introduces the concepts behind random and mixed effects models. -Christos Christos Hatzis, Ph.D. Vice President, Technology Nuvera Biosciences, Inc. 400 West Cummings Park Suite 5350 Woburn, MA 01801 Tel: 781-938-3830 www.nuverabio.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Crabb, David Sent: Tuesday, July 18, 2006 1:36 PM To: r-help@stat.math.ethz.ch Subject: [R] bilinear regression I think this is an easy question, but I would be grateful for any advice on how to implement this in R. I simply have a response variable (y) that I am trying to predict with one explanatory variable (x) but the shape of the scatter plot is distinctly bilinear. It would be best described by two straight lines. Is there a way of fitting a linear model to give me a bilinear fit and (more importantly) automatically determine the 'cut off' point? I would also want some statistic to convince myself that the bilinear fit is better. Many thanks for your help. David Dr. David Crabb Department of Optometry and Visual Science, City University, Northampton Square, London EC1V OHB Tel: 44 207 040 0191 [EMAIL PROTECTED] http://www.city.ac.uk/optometry/html/david_crabb.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Test for equality of coefficients in multivariate multiple regression
Hello,
suppose I have a multivariate multiple regression model such as the
following:
DF-data.frame(x1=rep(c(0,1),each=50),x2=rep(c(0,1),50))
tmp-rnorm(100)
DF$y1-tmp+DF$x1*.5+DF$x2*.3+rnorm(100,0,.5)
DF$y2-tmp+DF$x1*.5+DF$x2*.7+rnorm(100,0,.5)
x.mlm-lm(cbind(y1,y2)~x1+x2,data=DF)
coef(x.mlm)
y1y2
(Intercept) 0.07800993 0.2303557
x1 0.52936947 0.3728513
x2 0.13853332 0.4604842
How can I test whether x1 and x2 respectively have the same effect on y1
and y2? In other words, how can I test if coef(x.mlm)[2,1] is
statistically equal to coef(x.mlm)[2,2] and coef(x.mlm)[3,1] to
coef(x.mlm)[3,2]? I looked at linear.hypothesis {car} and glh.test
{gmodels}, but these do not seem the apply to multivariate models.
Thank you in advance,
Uli Keller
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] JGR help()
Dear R-helpers, In JGR, how to I get the help() to update when I install a new package? sessionInfo() Version 2.3.1 (2006-06-01) powerpc-apple-darwin8.6.0 attached base packages: [1] datasets methods stats graphics grDevices utils base other attached packages: lattice MASSJGR JavaGD rJava 0.13-9 7.2-27.11.4-20.3-30.4-3 _ Professor Michael Kubovy University of Virginia Department of Psychology USPS: P.O.Box 400400Charlottesville, VA 22904-4400 Parcels:Room 102Gilmer Hall McCormick RoadCharlottesville, VA 22903 Office:B011+1-434-982-4729 Lab:B019+1-434-982-4751 Fax:+1-434-982-4766 WWW:http://www.people.virginia.edu/~mk9y/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Using corStruct in nlme
I am having trouble fitting correlation structures within nlme. I would like to
fit corCAR1, corGaus and corExp correlation structures to my data. I either
get the error step halving reduced below minimum in pnls step or
alternatively R crashes.
My dataset is similar to the CO2 example in the nlme package. The one major
difference is that in my case the 'conc' steps are not the same for each
'Plant'. I have replicated the problem using the CO2 data in nlme (based off of
the Ch08.R script).
This works (when 'conc' is the same for each 'Plant':
(fm1CO2.lis - nlsList(SSasympOff, CO2))
(fm1CO2.nlme - nlme(fm1CO2.lis, control = list(tolerance = 1e-2)))
(fm2CO2.nlme - update(fm1CO2.nlme, random = Asym + lrc ~ 1))
CO2.nlme.var - update(fm2CO2.nlme,
fixed = list(Asym ~ Type * Treatment, lrc + c0 ~ 1),
start = c(32.412, 0, 0, 0, -4.5603, 49.344),
weights=varConstPower(fixed=list(const=0.1, power=1)), verbose=T)
CO2.nlme.CAR-update(CO2.nlme.var, corr=corCAR1())
CO2.nlme.gauss-update(CO2.nlme.var,
correlation=corGaus(form=~as.numeric(conc)|Plant,nugget=F), data=CO2)
CO2.nlme.exp-update(CO2.nlme.var,
correlation=corExp(form=~as.numeric(conc)|Plant,nugget=F), data=CO2)
But, if i change each of the 'conc' numbers slightly so that they are no longer
identical between subjects i can only get the corCAR1 correlation to work while
R crashes for both corExp and corGaus:
for(i in 1:length(CO2$conc)){
CO2$conc[i]-(CO2$conc[i]+rnorm(1))
}
(fm1CO2.lis - nlsList(SSasympOff, CO2))
(fm1CO2.nlme - nlme(fm1CO2.lis, control = list(tolerance = 1e-2)))
(fm2CO2.nlme - update(fm1CO2.nlme, random = Asym + lrc ~ 1))
CO2.nlme.var - update(fm2CO2.nlme,
fixed = list(Asym ~ Type * Treatment, lrc + c0 ~ 1),
start = c(32.412, 0, 0, 0, -4.5603, 49.344),
weights=varConstPower(fixed=list(const=0.1, power=1)), verbose=T)
CO2.nlme.CAR-update(CO2.nlme.var, corr=corCAR1())
CO2.nlme.gauss-update(CO2.nlme.var,
correlation=corGaus(form=~as.numeric(conc)|Plant,nugget=F), data=CO2)
CO2.nlme.exp-update(CO2.nlme.var,
correlation=corExp(form=~as.numeric(conc)|Plant,nugget=F), data=CO2)
I have read Pinheiro Bates (2000) and i think that it should be possible to
fit these correlation structures to my data, but maybe i am mistaken.
I am running R 2.3.1 and have recently updated all packages.
Thanks,
Katie Grieve
Quantitative Ecology Resource Management
University of Washington
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] R-help in a newsgroup
Hi, I find a lot of the R-help email traffic overloads my inbox. My IT managers are not really happy for me to be subscribed to several high-traffic email lists. I don't want to lose my contact with the R-help emails, so I'm having to consider various ways of handling the traffic. Anyhow, I'm wondering how many people on the R-help email list would prefer that most of the traffic were in a newsgroup? In case your interested in that option, there is a group available at: [EMAIL PROTECTED] I think the subscription is through normal news group channels. The google search services on this group are nice too. This group is not divided into useful categories, like help, admin, develop etc., but it's not too difficult to create new groups for that. Best, Darren __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Reproducible Research - Examples
All, Recently I ran across a URL documenting published research using R: http://www.cgd.ucar.edu/ccr/ammann/millennium/CODES_MBH.html A note on the site indicates that the code is being revised. The code and data are provided, so that one could reproduce the results without having to buy a proprietary software program. In poking around the R website it is clear that a lot of thought has gone into documenting reproducible research, notably by Harrell, Gentleman, and the Sweave effort, among others. http://biostat.mc.vanderbilt.edu/twiki/bin/view/Main/StatReport http://cran.ssds.ucdavis.edu/doc/contrib/Harrell-statcomp-notes.pdf http://www.bioconductor.org/docs/papers/2003/Compendium Question: Reproducible research is clearly desirable and feasible. Could anyone provide examples (stand-alone URLs or supplementary material in journals) that you would recommend as a models for reproducible research in R? Thanks, Phil Philip Heilman USDA-ARS Southwest Watershed Research Center 2000 E. Allen Rd. Tucson, AZ 85704 (520) 670-6380 x148 [EMAIL PROTECTED] http://www.tucson.ars.ag.gov/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] R-help in a newsgroup
On Tue, 2006-07-18 at 11:30 -0700, Darren Weber wrote: Hi, I find a lot of the R-help email traffic overloads my inbox. My IT managers are not really happy for me to be subscribed to several high-traffic email lists. I don't want to lose my contact with the R-help emails, so I'm having to consider various ways of handling the traffic. Anyhow, I'm wondering how many people on the R-help email list would prefer that most of the traffic were in a newsgroup? In case your interested in that option, there is a group available at: [EMAIL PROTECTED] I think the subscription is through normal news group channels. The google search services on this group are nice too. This group is not divided into useful categories, like help, admin, develop etc., but it's not too difficult to create new groups for that. Best, Darren r-help is already available with an NNTP interface at gmane.org: http://dir.gmane.org/gmane.comp.lang.r.general There is also a web based interface, where you can see that your post is already available: http://news.gmane.org/gmane.comp.lang.r.general Similarly, r-devel is also present: http://dir.gmane.org/gmane.comp.lang.r.devel The Google group you reference is completely independent of the R e-mail lists, whereas the gmane interface is synchronized with the R e-mail lists. HTH, Marc Schwartz __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] R-help in a newsgroup
On 7/18/2006 2:30 PM, Darren Weber wrote: Hi, I find a lot of the R-help email traffic overloads my inbox. My IT managers are not really happy for me to be subscribed to several high-traffic email lists. I don't want to lose my contact with the R-help emails, so I'm having to consider various ways of handling the traffic. Anyhow, I'm wondering how many people on the R-help email list would prefer that most of the traffic were in a newsgroup? In case your interested in that option, there is a group available at: [EMAIL PROTECTED] I think the subscription is through normal news group channels. The google search services on this group are nice too. This group is not divided into useful categories, like help, admin, develop etc., but it's not too difficult to create new groups for that. If you prefer the newsgroup interface, you should also look at gmane. Gabor G posted a list of the newsgroups here: http://finzi.psych.upenn.edu/R/Rhelp02a/archive/75239.html recently. Duncan Murdoch Best, Darren __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Question about summing to zero
Hi, When I used x - c(10,10,10,10,5,5,5,5) sum(x) x - x - .Last.value/8 the result was zero, as I expected. However, when I used x - rnorm(101,0,10) sum(x) x - x - .Last.value/101 sum(x) I did not get zero, but -2.664535e-15. OK, that's fairly close to zero, but I tried to improve the figure. Each time I repeated the last two commands, the sum got closer to zero. The series seemed to converge on -1.30e-15, so changed the 101 to 50, then to 20, and then to 10. I had sum(x) down to -5.551115e-17 before I quit. If R can store values of x which sum to -5.551115e-17, then why can't I get that value on the first try? Is this a bug or just rounding error? Jim Frederick __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test for equality of coefficients in multivariate multiple regression
Dear Ulrich,
I'll look into generalizing linear.hypothesis() so that it handles
multivariate linear models.
Meanwhile, vcov(x.mlm) will give you the covariance matrix of the
coefficients, so you could construct your own test by ravelling
coef(x.mlm) into a vector.
I hope that this helps,
John
On Tue, 18 Jul 2006 20:15:12 +0200
Ulrich Keller [EMAIL PROTECTED] wrote:
Hello,
suppose I have a multivariate multiple regression model such as the
following:
DF-data.frame(x1=rep(c(0,1),each=50),x2=rep(c(0,1),50))
tmp-rnorm(100)
DF$y1-tmp+DF$x1*.5+DF$x2*.3+rnorm(100,0,.5)
DF$y2-tmp+DF$x1*.5+DF$x2*.7+rnorm(100,0,.5)
x.mlm-lm(cbind(y1,y2)~x1+x2,data=DF)
coef(x.mlm)
y1y2
(Intercept) 0.07800993 0.2303557
x1 0.52936947 0.3728513
x2 0.13853332 0.4604842
How can I test whether x1 and x2 respectively have the same effect on
y1
and y2? In other words, how can I test if coef(x.mlm)[2,1] is
statistically equal to coef(x.mlm)[2,2] and coef(x.mlm)[3,1] to
coef(x.mlm)[3,2]? I looked at linear.hypothesis {car} and glh.test
{gmodels}, but these do not seem the apply to multivariate models.
Thank you in advance,
Uli Keller
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario, Canada
http://socserv.mcmaster.ca/jfox/
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Question about summing to zero
On Tue, 18 Jul 2006, Jim Frederick wrote:
Hi,
When I used
x - c(10,10,10,10,5,5,5,5)
sum(x)
x - x - .Last.value/8
the result was zero, as I expected.
However, when I used
x - rnorm(101,0,10)
sum(x)
x - x - .Last.value/101
sum(x)
I did not get zero, but -2.664535e-15. OK, that's fairly close to
zero, but I tried to improve the figure. Each time I repeated the last
two commands, the sum got closer to zero. The series seemed to converge
on -1.30e-15, so changed the 101 to 50, then to 20, and then to 10. I
had sum(x) down to -5.551115e-17 before I quit.
If R can store values of x which sum to -5.551115e-17, then why can't I
get that value on the first try? Is this a bug or just rounding error?
It's just rounding error. The *relative* error is always a multiple of
2^-52, or about 2e-16 but as sum(x) gets smaller a given relative error is
a smaller absolute error.
Comparing values from doing this once or twice is not very reliable. To
see how much variation there is from sample to sample, do something like
a-replicate(1, {x-rnorm(101,0,10); y-sum(x);
sum(x-y/101)/.Machine$double.eps})
A histogram or qqplot shows an approximately Normal distribution, with
mean close to zero and standard deviation about 70. Zero error does
happen: I got 20/1 exactly zero.
There aren't any simple answers to exactly how big the error is for
particular computations. The exact answers are complicated and the simple
answers are approximate. For example, in your case a reasonable upper
bound for the error might be 102 computations multiplied by 10 as a
typical value of the numbers, multiplied by .Machine$double.eps. This is
much larger than the typical error, of course. A crude estimate of the
typical error might be sqrt(102)*10*.Machine$double.eps, treating the
errors as a random walk, and this is not far off (though perhaps only
coincidentally so).
-thomas
Thomas Lumley Assoc. Professor, Biostatistics
[EMAIL PROTECTED] University of Washington, Seattle
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] Extended example of R2HTML?
Dear R-helpers, Can someone point me to an extended example of use of R2HTML? _ Professor Michael Kubovy University of Virginia Department of Psychology USPS: P.O.Box 400400Charlottesville, VA 22904-4400 Parcels:Room 102Gilmer Hall McCormick RoadCharlottesville, VA 22903 Office:B011+1-434-982-4729 Lab:B019+1-434-982-4751 Fax:+1-434-982-4766 WWW:http://www.people.virginia.edu/~mk9y/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] FW: Large datasets in R
Well, SPSS used to claim that all its algorithms dealt with only one case at a time and therefore that it could handle very large files. I suppose a large correlation matrix could cause it problems. Marsh Feldmman -Original Message- From: Ritwik Sinha [mailto:[EMAIL PROTECTED] Sent: Tuesday, July 18, 2006 10:54 AM To: Prof Brian Ripley Cc: Marshall Feldman; r-help@stat.math.ethz.ch Subject: Re: [R] FW: Large datasets in R Hi, I have a related question. How differently do other statistical softwares handle large data? The original post claims that 350 MB is fine on Stata. Some one suggested S-Plus. I have heard people say that SAS can handle large data sets. Why can others do it and R seem to have a problem? Don't these softwares load the data onto RAM. -- Ritwik Sinha Graduate Student Epidemiology and Biostatistics Case Western Reserve University http://darwin.cwru.edu/~rsinha __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] How to write a function in a graph
Dear R-ers, I conducted a regression analysis, and then intended to add the regression function (y=4.33+1.07x) in a graph. But the following code can only give me a text like y=a+bx. Who can help me out? Thank you very much in advance. CODE # Read data x - c(1, 2, 3, 4, 5, 6, 7, 8, 9) y - c(6, 5, 8, 9, 11, 10, 11, 12, 15) data01 - data.frame(x, y) # Regression analysis res.lm.y - nls(y~a+b*x, start=list(a=1, b=2),data=data01) # Obtain parameters a- coef(res.lm.y)[a] b- coef(res.lm.y)[b] a b #a=4.33 #b=1.07 # Plot the results def.par - par() par(mfrow=c(1,1),xaxs=i,yaxs=i) plot(data01$x,data01$y,main=Fit,xlab=x,ylab=y) lines(data01$x,predict(res.lm.y)) #=== text (6, 13, expression(y==a+b*x)) #=== ## I intended to add text like y=4.33+1.07x ## but the above code added y=a+bx Swiss Federal Institute for Environmental Science and Technology (EAWAG) Ueberlandstrasse 133 P.O.Box 611 CH-8600 Duebendorf Switzerland Phone: 0041-18235012 Fax: 0041-18235375 - [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] FW: Large datasets in R
[Thomas Lumley] People have used R in this way, storing data in a database and reading it as required. There are also some efforts to provide facilities to support this sort of programming (such as the current project funded by Google Summer of Code: http://tolstoy.newcastle.edu.au/R/devel/06/05/5525.html). Interesting project indeed! However, if R requires uses more swapping because arrays do not all fit in physical memory, crudely replacing swapping with database accesses is not necessarily going to buy a drastic speed improvement: the paging gets done in user space instead of being done in the kernel. Long ago, while working on CDC mainframes, astonishing at the time but tiny by nowadays standards, there was a program able to invert or do simplexes on very big matrices. I do not remember the name of the program, and never studied it but superficially (I was in computer support for researchers, but not a researcher myself). The program was documented as being extremely careful at organising accesses to rows and columns (or parts thereof) in such a way that real memory was best used. In other words, at the core of this program was a paging system very specialised and cooperative with the problems meant to be solved. However, the source of this program was just plain huge (let's say from memory, about three or four times the size of the optimizing FORTRAN compiler, which I already knew better as an impressive algorithmic undertaking). So, good or wrong, the prejudice stuck solidly in me at the time, if nothing else, that handling big arrays the right way, speed-wise, ought to be very difficult. One reason there isn't more of this is that relying on Moore's Law has worked very well over the years. On the other hand, the computational needs for scientific problems grow fairly quickly to the size of our ability to solve them. Let me take weather forecasting for example. 3-D geographical grids are never fine enough for the resolution meteorologists would like to get, and the time required for each prediction step grows very rapidly, to increase precision by not so much. By merely tuning a few parameters, these people may easily pump nearly all the available cycles out the supercomputers given to them, and they do so without hesitation. Moore's Law will never succeed at calming their starving hunger! :-). -- François Pinard http://pinard.progiciels-bpi.ca __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Plot fit of a generic function
Hello!
Say I have a function, which creates a design matrix i.e.
myFunc - function(x)
{
ret - cbind(x, x*x, x*x*x)
colnames(ret) - 1:ncol(ret)
return(ret)
}
n - 200
x - runif(n=n, min=0, max=100)
y - myFunc(x) %*% c(1, 0.2, -0.0002) + rnorm(n=n, sd=100)
then I can use this in formulae as here
(fit - lm(y ~ myFunc(x)))
Now I would like to plot data and fitted function on the plot, but I do
not want to access each parameter estimate from object fit i.e. I
would like to use something similar to abline for linear regression but
in a generic way. Is there anything similar to my case?
plot(y=y, x=x)
???plotMyFunc???
Thanks!
--
Lep pozdrav / With regards,
Gregor Gorjanc
--
University of Ljubljana PhD student
Biotechnical Faculty
Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan
Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si
SI-1230 Domzale tel: +386 (0)1 72 17 861
Slovenia, Europefax: +386 (0)1 72 17 888
--
One must learn by doing the thing; for though you think you know it,
you have no certainty until you try. Sophocles ~ 450 B.C.
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] FW: Large datasets in R
Or, more succinctly, Pinard's Law: The demands of ever more data always exceed the capabilities of ever better hardware. ;-D -- Bert Gunter Genentech Non-Clinical Statistics South San Francisco, CA -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of François Pinard Sent: Tuesday, July 18, 2006 3:56 PM To: Thomas Lumley Cc: r-help@stat.math.ethz.ch Subject: Re: [R] FW: Large datasets in R [Thomas Lumley] People have used R in this way, storing data in a database and reading it as required. There are also some efforts to provide facilities to support this sort of programming (such as the current project funded by Google Summer of Code: http://tolstoy.newcastle.edu.au/R/devel/06/05/5525.html). Interesting project indeed! However, if R requires uses more swapping because arrays do not all fit in physical memory, crudely replacing swapping with database accesses is not necessarily going to buy a drastic speed improvement: the paging gets done in user space instead of being done in the kernel. Long ago, while working on CDC mainframes, astonishing at the time but tiny by nowadays standards, there was a program able to invert or do simplexes on very big matrices. I do not remember the name of the program, and never studied it but superficially (I was in computer support for researchers, but not a researcher myself). The program was documented as being extremely careful at organising accesses to rows and columns (or parts thereof) in such a way that real memory was best used. In other words, at the core of this program was a paging system very specialised and cooperative with the problems meant to be solved. However, the source of this program was just plain huge (let's say from memory, about three or four times the size of the optimizing FORTRAN compiler, which I already knew better as an impressive algorithmic undertaking). So, good or wrong, the prejudice stuck solidly in me at the time, if nothing else, that handling big arrays the right way, speed-wise, ought to be very difficult. One reason there isn't more of this is that relying on Moore's Law has worked very well over the years. On the other hand, the computational needs for scientific problems grow fairly quickly to the size of our ability to solve them. Let me take weather forecasting for example. 3-D geographical grids are never fine enough for the resolution meteorologists would like to get, and the time required for each prediction step grows very rapidly, to increase precision by not so much. By merely tuning a few parameters, these people may easily pump nearly all the available cycles out the supercomputers given to them, and they do so without hesitation. Moore's Law will never succeed at calming their starving hunger! :-). -- François Pinard http://pinard.progiciels-bpi.ca __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Plot fit of a generic function
You could plot y vs. fitted(y.lm) where y.lm is the output of lm or
plot both y and fitted(y.lm) against x on the same chart.
On 7/18/06, Gregor Gorjanc [EMAIL PROTECTED] wrote:
Hello!
Say I have a function, which creates a design matrix i.e.
myFunc - function(x)
{
ret - cbind(x, x*x, x*x*x)
colnames(ret) - 1:ncol(ret)
return(ret)
}
n - 200
x - runif(n=n, min=0, max=100)
y - myFunc(x) %*% c(1, 0.2, -0.0002) + rnorm(n=n, sd=100)
then I can use this in formulae as here
(fit - lm(y ~ myFunc(x)))
Now I would like to plot data and fitted function on the plot, but I do
not want to access each parameter estimate from object fit i.e. I
would like to use something similar to abline for linear regression but
in a generic way. Is there anything similar to my case?
plot(y=y, x=x)
???plotMyFunc???
Thanks!
--
Lep pozdrav / With regards,
Gregor Gorjanc
--
University of Ljubljana PhD student
Biotechnical Faculty
Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan
Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si
SI-1230 Domzale tel: +386 (0)1 72 17 861
Slovenia, Europefax: +386 (0)1 72 17 888
--
One must learn by doing the thing; for though you think you know it,
you have no certainty until you try. Sophocles ~ 450 B.C.
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] JavaGD
Can someone please help im trying to use JGR' version 1.4-2 on a IBM Thinkcentre P4 with windows xp with the R v2.3.1, but any time I try and plot or use JavaGD I get this error plot(1:10) Error in JavaGD() : unable to start device JavaGD In addition: Warning message: Another VM is running already and the VM did not allow me to append paths to the class path. in: .jinit(.javaGD.get.class.path()) Or JavaGD(name=JavaGD, width=400, height=300, ps=12) Error in JavaGD(name = JavaGD, width = 400, height = 300, ps = 12) : unable to start device JavaGD if I try and reinstall I get Warning: package 'JavaGD' is in use and will not be installed Any help will be greatly appreciated Regards Mark Castalanelli Bsc (Hons) Molecular Biology Bsc Forensic Biology and Toxicology DAFWA - European House Borer Project (08) 9368 3754 This e-mail and any files transmitted with it are privileged and confidential information intended for use of the addressee.The confidentiality and/or privilege is not waived, lost or destroyed if it has been transmitted to you in error. If you received this e-mail in error you must (a) not disseminate, copy or take any action in reliance on it; (b) please notify the Department of Agriculture immediately by return e-mail to the sender; (c) please delete the original e-mail. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test for equality of coefficients in multivariate multiple regression
Hi Uli,
I suggest that you try to rewrite the model system into a single
mixed-effects model, which would allow direct parameterization of the
tests that you're interested in. A useful article, which may be
overkill for your needs, is:
Hall, D.B. and Clutter, M. (2004). Multivariate multilevel nonlinear
mixed effects models for timber yield predictions, Biometrics, 60:
16-24.
See the publications link on Daniel Hall's website:
http://www.stat.uga.edu/~dhall/
Cheers
Andrew
On Tue, Jul 18, 2006 at 08:15:12PM +0200, Ulrich Keller wrote:
Hello,
suppose I have a multivariate multiple regression model such as the
following:
DF-data.frame(x1=rep(c(0,1),each=50),x2=rep(c(0,1),50))
tmp-rnorm(100)
DF$y1-tmp+DF$x1*.5+DF$x2*.3+rnorm(100,0,.5)
DF$y2-tmp+DF$x1*.5+DF$x2*.7+rnorm(100,0,.5)
x.mlm-lm(cbind(y1,y2)~x1+x2,data=DF)
coef(x.mlm)
y1y2
(Intercept) 0.07800993 0.2303557
x1 0.52936947 0.3728513
x2 0.13853332 0.4604842
How can I test whether x1 and x2 respectively have the same effect on y1
and y2? In other words, how can I test if coef(x.mlm)[2,1] is
statistically equal to coef(x.mlm)[2,2] and coef(x.mlm)[3,1] to
coef(x.mlm)[3,2]? I looked at linear.hypothesis {car} and glh.test
{gmodels}, but these do not seem the apply to multivariate models.
Thank you in advance,
Uli Keller
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
--
Andrew Robinson
Department of Mathematics and StatisticsTel: +61-3-8344-9763
University of Melbourne, VIC 3010 Australia Fax: +61-3-8344-4599
Email: [EMAIL PROTECTED] http://www.ms.unimelb.edu.au
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] [R-pkgs] odfWeave Package
I've been meaning send an announcement for this package, but Greg Snow
beat me to the punch today.
Max
snip
The odfWeave package is now available on CRAN at
http://lib.stat.cmu.edu/R/CRAN/src/contrib/Descriptions/odfWeave.html
and your local mirror.
The package extends Sweave to Open Document Format (ODF) text document
files. Latex-style code chunks and in-line Sexpr commands can be used
to embed R output into ODF file, which can then be exported to doc,
rtf, html, pdf and other formats.
Other functions in the package facilitate ODF formatted tables, text
and lists.
The current limitations of the package are:
O tangling is not yet implement (but will be)
O testing has been done on text documents generated by OpenOffice.
Other formats/editors may work, but are not (yet) supported.
Please send comments, suggestions and/or contributions.
Max Kuhn
Research Statistics
Pfizer Global RD
Max.Kuhn at pfizer.com
--
LEGAL NOTICE\ Unless expressly stated otherwise, this messag...{{dropped}}
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] How to find S4 generics? (was: inames() function and lmer())
* * methods * * You have asked an excellent question. I can provide a partial answer below. First, however, I wish to pose a question of my own, which could help answer your question: How can one obtain a simple list of the available generics for a class? For an S3 class, the 'methods' functions provide that. What about an S4 class? That's entirely opaque to me, if I somehow can't find the relevant information in other ways. For example, ?lmer-class lists many but not all of the methods available for objects of class 'lmer'. I think I once found a way to get that, but I'm not able to find documentation on it now. * * Retrieving information from an 'lmer' object * * There are many ways to retrieve information from an object. For 'lmer', I think the best source of information is probably the help pages for 'lmer' and 'lmer-class'. The latter tells you that there are Methods anova, coef, coerce, deviance, logLik, update, simulate, solve, terms, vcov, ranef and VarCorr for extracting information or modifying an object of class 'lmer'. There are also other methods not listed on that page like 'fixef'. I don't know how to find those easily. There is also the excellent vignette(MlmSoftRev) in the 'mlmRev' package. (For help with it, see, e.g., http://finzi.psych.upenn.edu/R/Rhelp02a/archive/76134.html;.) Finally, the function 'str' will in most cases expose the structure of an object, thereby making it relatively easy to figure out how to get what is needed for many applications. That does not always work, however, because 'str' is a generic function, not a primitive. 'str' seems to expose everything one might possibly want to know about an 'lmer' object (apart from the theory behind it). This is not the case for an object of class 'logLik', because a function 'str.logLik' provides a succinct summary of a 'logLik' object in a non-standard format. The following will defeat 'str.logLik': fm1 - lmer(Reaction ~ Days + (Days|Subject), sleepstudy) lglk1 - logLik(fm1) str(lglk1) Class 'logLik' : -871.8 (df=5) # I don't understand this. str(unclass(lglk1)) # This is in the standard 'str' format: atomic [1:1] -872 - attr(*, nobs)= int 180 - attr(*, nall)= int 180 - attr(*, df)= num 5 - attr(*, REML)= logi TRUE Best Wishes, Spencer Graves A.R. Criswell wrote: Hello All, I would like to retrieve some of the results from the lmer(...) function in library lme4. If I run a model, say fm.1 - lmer(y ~ 1 + (1 | x), data = dog) and try names(fm.1), I get NULL. Is there anyway to retrieve the information? Thanks __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] voronoi tessellations
Hello, looking to draw a voronoi tessellations in R - can anyone recommend a package that has tackled this? some background: i have a economic data set and created a sammons projection, like to now overlay a voronoi tessellation over the sammons 2-D solution for a slick visual, and potentially color each tessellation element based on a metric. home.u - unique(home1) home.dist - dist(home.u) home.sam - sammon(home.dist,k=2) plot(home.sam$points) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Problem with ordered logistic regression using polr function.
Hi, I'm trying to fit a ordered logistic regression. The response variable (y) has three levels (0,1,2). The command I've used is: /ordlog-polr(y~x1+x2+x3+x4, data=finalbase, subset=heard, weight=wt, na.action=na.omit) / (There are no NA's in y but there are NA's in X's) The error I'm getting is: /Warning messages: 1: non-integer #successes in a binomial glm! in: eval(expr, envir, enclos) 2: design appears to be rank-deficient, so dropping some coefs in: polr(right.ans ~ S107 + children + work + rel + media + V013 + / Also, if I write /summary(ordlog)/ I'm getting the following error: /Re-fitting to get Hessian Error in if (all(pr 0)) -sum(wt * log(pr)) else Inf : missing value where TRUE/FALSE needed /Could anybody point out the problem? Thanks, Debarchana. -- Debarchana Ghosh Research Assistant Department of Geography University of Minnesota PH: 8143607580 email to: [EMAIL PROTECTED] www.tc.umn.edu/~ghos0033 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Problem with ordered logistic regression using polr function
Hi, I'm trying to fit a ordered logistic regression. The response variable (y) has three levels (0,1,2). The command I've used is: ordlog-polr(y~x1+x2+x3+x4, data=finalbase, subset=heard, weight=wt, na.action=na.omit) (There are no NA's in y but there are NA's in X's) The error I'm getting is: Warning messages: 1: non-integer #successes in a binomial glm! in: eval(expr, envir, enclos) 2: design appears to be rank-deficient, so dropping some coefs in: polr(right.ans ~ S107 + children + work + rel + media + V013 + Also, if I write summary(ordlog) I'm getting the following error: Re-fitting to get Hessian Error in if (all(pr 0)) -sum(wt * log(pr)) else Inf : missing value where TRUE/FALSE needed Could anybody point out the problem? Thanks, Debarchana. -- Debarchana Ghosh Research Assistant Department of Geography University of Minnesota PH: 8143607580 email to: [EMAIL PROTECTED] www.tc.umn.edu/~ghos0033 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] voronoi tessellations
I'll suggest going to the CRAN packages page and doing a search for voronoi. Also, search for 'triangulation', since that is one of the uses of them. -Don At 11:46 PM -0400 7/18/06, zubin wrote: Hello, looking to draw a voronoi tessellations in R - can anyone recommend a package that has tackled this? some background: i have a economic data set and created a sammons projection, like to now overlay a voronoi tessellation over the sammons 2-D solution for a slick visual, and potentially color each tessellation element based on a metric. home.u - unique(home1) home.dist - dist(home.u) home.sam - sammon(home.dist,k=2) plot(home.sam$points) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- - Don MacQueen Lawrence Livermore National Laboratory Livermore, CA, USA __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] conditional plot
Hi, Can anyone pls help me in plotting the following data? The data-set contains company name, specification-1, specification-2. The graph would basically plot company name with specification-1 on x-axis, specification-2 on y-axis. Simply put - company name should get classified into one of the four quardrants created by specification 1 specification2. Thanks. Manoj __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] ts and stl functions - still a problem
Have you tried the following: tstkr - ts(tkr[, 1], deltat=1/12) I had previously thought that 'tkr' was a numeric matrix; if that had been the case, as.numeric(tkr) would have converted it to a numeric vector. You've now said that class(tkr) = 'data.frame' of dim = c(131, 1). If this is the case, then tkr[, 1] should be a (numeric) vector. This should solve your problem. However, I encourage you to spend more time studying the similarities and differences between vectors, matrices and data.frames, especially the examples in the ?matrix and ?data.frame help pages, plus the corresponding sections of An Introduction to R, the first of the Manuals available via 'help.start()'. Hope this helps. Spencer Graves Daniel sutcliffe wrote: I am still having a problem with my coercing my data frame (tkr) into a time series using ts to use the stl function. When I read the data into R and do dim tkr the result is 132 1, when I type class I get (data.frame). I have tried several methods so far to to coerce it into a a time series and use the stl function,( see thread) . A suggestion was to use the following: tstkr - ts(as.numeric(tkr), deltat=1/12) before using the stl function, however I now get this error message: Error in as.double.default(tkr) : (list) object cannot be coerced to 'double' I don't know what this meansdoes anybody have any other ideas...? Hope I have put enough information on here for people to be able to help..if not let me know, also if anyone wants to send me to email them my data then I am happy to do so.. Thanks to everyone who has tried to help so far much appreciated... Spencer Graves [EMAIL PROTECTED] wrote: The 'ts' function retains the 'dim' attribute of 'tkr'. When 'stl' finds this 'dim' attribute, it thinks 'tstkr' is a multivariate time series. This causes it to stop with the error, only univariate series allowed. Consider the following modification of an example from the 'stl' help file: not.1 - stl(nottem, per) Nottem - ts(array(nottem, dim=c(240, 1))) Not.1 - stl(Nottem) Error in stl(Nottem) : only univariate series are allowed Solution: tstkr - ts(as.numeric(tkr), deltat=1/12) After converting tkr and tstkr from a matrix to a vector like this, please try 'stl(tstkr)'. If it doesn't work, please submit another post. Hope this helps. Spencer Graves p.s. Your example was not quite self-contained, because I didn't know for sure the format, class, and attributes of your 'tkr' object. The absence of these details made it harder for me (and, I believe, anyone else) to reply. You might get better replies quicker with greater attention to such details. Daniel sutcliffe wrote: Hi I am still having problems with using the stl function, when I read the csv file into R into a file called tkr and use dim(tkr) the result is 132 x 1 which is fine. When coerce it into a trime series using ts either: tstkr - ts(t(tkr), deltat=1/12) or tstkr - ts(c(tkr), deltat=1/12) and use the stl function I get the following error: Error in stl(tstkr) : only univariate series are allowed id just use the tkr file I get the same error..does anyibe have an idea what to do next, here is my data...it's not sensitive so if anyone wants to try then you are very welcome! Rate 184.0222 180.517 222.5792 173.5066 192.7852 198.0429 182.2696 189.28 178.7644 206.8059 236.6 155.9807 231.7314 249.2868 222.9537 198.3761 201.8872 208.9094 242.2646 221.1982 228.2203 245.7757 244.0202 194.865 239.3664 234.0862 251.6867 235.8463 197.1253 237.6063 267.5271 228.8061 241.1264 249.9267 256.9669 188.325 258.8788 239.5069 214.8518 234.2237 211.3296 234.2237 237.7458 156.7361 239.5069 225.4183 257.1177 170.8248 230.1611 265.3001 296.9253 193.265 233.675 240.7028 249.4876 205.5637 237.1889 237.1889 289.8975 245.9736 283.1755 316.3875 372.3234 234.2316 263.9476 314.6395 286.6715 272.6875 323.3795 295.4115 300.6555 174.7997 227.184 277.4767 317.364 234.121 286.1478 279.2109 280.9452 235.8552 319.0982 296.5532 303.4901 173.4229 302.6072 312.8651 389.7991 223.9635 254.7371 329.9615 288.93 317.994 312.8651 381.2509 333.3808 194.8996 285.5743 304.0526 335.9698 307.4123 346.0489 288.934 335.9698 243.5781 384.6854 359.4876 357.8078 221.74 336.3712 344.6562 389.3952 286.6611 338.0282 407.6222 356.2552 241.9221 347.9702 400.9942 381.1102 304.8882 383.9077 418.5089 476.1774 331.1822 378.9647 375.6694 339.4206 364.1357 420.1565 446.5193 410.2705 296.5811 Cheers and thanks to everyone who offered suggestions before. Daniel SAULEAU Erik-André wrote: Perhaps ts(t(tkr))? -Message d'origine- De : Daniel sutcliffe [mailto:[EMAIL PROTECTED] Envoyé : mercredi 12 juillet 2006 15:53 À : r-help@stat.math.ethz.ch Objet : [R] ts and stl functions Hi, I have imported a csv
