Re: [R] read.spss and umlaut
Hello Am Donnerstag, 3. August 2006 15.34 schrieb Thomas Lumley: On Thu, 3 Aug 2006, Thomas Kuster wrote: Hello Am Mittwoch, 2. August 2006 17.11 schrieb Thomas Lumley: ... You haven't shown anything that indicates that the C code stopped reading. More likely R just stops displaying when it gets to an illegal byte sequence. You could use nchar() to count the bytes in the string to find out. If I change the translatable characters (overwrite the 0 between :#@'= and ~000 with ÄÖÜäöü). I can read in the file an every ÄÖÜäöü ist a withspace: daten - read.spss(projets_umlaut.por) levels(daten$PROJETX) [1] Bg Stammzellenforschung [2] Bb ber eine neue Finanzordnung [3] Bb Neugestaltung des Finanzausgleichs [4] nderrung Bg EOG Mutterschafturlaub [5] EV Postdienste f r alle [6] Bb ber B rgerrechtserwerb 3. Generation [7] Bb ber erleichterte Einb rung 2. Generation [8] Bg Steuerpaket . . . levels(daten$PROJETX)[208] [1] EV Gleiche Rechte f r Mann und Frau Gegenvorschlag charToRaw(levels(daten$PROJETX)[208]) [1] 45 56 20 47 6c 65 69 63 68 65 20 52 65 63 68 74 65 20 66 20 72 20 4d 61 6e [26] 6e 20 75 6e 64 20 46 72 61 75 20 47 65 67 65 6e 76 6f 72 73 63 68 6c 61 67 without change the table I get: daten - read.spss(projets.por) charToRaw(levels(daten$PROJETX)[208]) [1] 45 56 20 47 6c 65 69 63 68 65 20 52 65 63 68 74 65 20 66 The SPSS file is from: http://voxit.sidos.ch/update.asp?lang=d - Download der kumulierten Dateien Version 2.0 You must accept this: The Standardized Post-Vote Surveys: http://voxit.sidos.ch/agreement.asp?lang=emenu=0 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] gnlsControl
Daniel Coleman coleman.daniel at gene.com writes: When I run gnls I get the error: Error in nls(y ~ cbind(1, 1/(1 + exp((xmid - x)/exp(lscal, data = xy, : step factor 0.000488281 reduced below 'minFactor' of 0.000976563 My first thought was to decrease minFactor but gnlsControl does not contain minFactor nor nlsMinFactor (see below). It does however contain nlsMaxIter and nlsTol which I assume are the analogs of maxiter and tol in nls.control. I would be happy to hear from anyone who has an idea on what parameters in gnlsControl to change to get convergence. Try nlsTol with a large value, e.g. 0.3. If I am really desparate, I put the gnls in a try() structure, halving nlsTol until it fails, and take the last successful. 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] Questions about sweave...
Brian Lunergan ff809 at ncf.ca writes:
Evening all:
I'm taking a little time to experiment with R, Sweave, and Miktex/LaTex but
I've run up against some problems and -well- I hope that there are some on
the list who might have some suggestions. This will be kind of wordy as I
will include the complete files involved as I'm just not sure what I'm
looking for. Apologies at the outset.
This is Windows special
See A.12:
http://www.ci.tuwien.ac.at/~leisch/Sweave/FAQ.html
And put a commented (!sic) \usepackage into the header (don't remember where I
found this)
% \usepackage{Sweave}
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.
[R] geodesic distance (solution)
Hi,
has anyone ever seen implemented in R the following geodesic
distance between positive definite pxp matrices A and B?
d(A,B) = \sum_{i=1}^p (\log \lambda_i)^2
were \lambda is the solution of det(A -\lambda B) = 0
thanks
stefano
as I received few private email on the claimed solution, I'm posting
it to r-help.
when matrix B is invertible (which is always my case), one approach
is to notice that
solving
det(A -\lambda * B) = 0
is equivalent to solve
det(B^-1*A -\lambda *I) = 0
which is a standard eigen value problem for the matrix B^-1 * A, hence
eigen(solve(B) %*% A)$values
is the answer.
I'm pretty sure that the problem can also be solved using some svd
decomposition when B is not invertible.
hope it helps
stefano
__
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] Building a random walk vector
Matthew Wilson matt at overlook.homelinux.net writes: I'm studying R in my free time. I want to build a vector where each element is equal to the element before it in the sequence plus some random tweak. You will probably get many answers to this, but I think vec - 100+c(0,cumsum(runif(49,-2,2))) works. 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.
[R] need sample parallelized R scripts
Good day to everyone. I'm working on computing correlation for several datasets (one dataset for each chromosome). Computation is done several thousand times for each dataset which at present takes around 13 hours. We have a HPC machine with MPI. snow package and R 2.3.1 running in Linux (Rocks) are installed in the machine. I need to modify the script to run it on several nodes. I do not have experience in writing parallel scripts. Anybody knows where I can find sample R scripts that are designed to run on parallel machines? Thanks. Ems Emily Deomano Crop Research Informatics Laboratory The International Rice Research Institute DAPO Box Metro Manila, Philippines Email: [EMAIL PROTECTED]; Tel No: (63-2) 580-5600; Fax No: (63-2) 580-5699 __ 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] Integration and Loop in R
Dear All,
I have seldom needed to use loops in R, but now I need to code a loop
with a stride different from one.
In the R manual I downloaded I have the example:
xc - split(x, ind)
yc - split(y, ind)
for (i in 1:length(yc)) {
plot(xc[[i]], yc[[i]]);
abline(lsfit(xc[[i]], yc[[i]]))
}
but in my case I'd like to add a condition so that i varies by 4 from
one go to the following one. I cannot figure out the right syntax, can
anyone help here?
Another thing (which could possibly solve my problem): I had a look at
integrate command in R.
It seems to require an object defined as a function to carry out the
integration.
What if I simply have a list of data values? How can I coerce them
into a function recognized by R? Furthermore, are there R routines to
carry out the integration on a non-equally spaced 1D grid?
Best Regards
Lorenzo
__
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] bullseye or polar display of circular data
Michael Jerosch-Herold wrote: I have data for several rings of a left heart chamber, and which I would like to display in concentric rings, with color-encoding of the values. Each ring corresponds to one slice through the heart, and the rings correspond to positions from the base to the apex of the heart as you move from the outermost ring to the innermost one. The data have a circular pattern. These types of displays are referred to as bullseye displays in the nuclear medicine literature. Does any reader of these messages know of a R function/package that offers this functionality? Also I noticed that in some contexts you can define a circular attribute for your data. Are there plot routines for such circular data? I'm not quite sure that I understand the display you want, but radial.plot in the plotrix package might do what you want. Also you could check the circular and CircStats packages. Jim __ 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] prettyR arrives
Hi all, I have finally gotten the prettyR package going (many thanks to Kurt Hornik for his patience). prettyR is a set of functions that allows the user to produce HTML output from R scripts. Given an R script that runs properly, an HTML listing complete with embedded graphics can be produced simply by passing the script to the core function htmlize (Phillipe Grosjean has not only offered great suggestions, but provided a fancier function named R2html). It is even possible to have the output magically appear in your friendly local HTML browser when the script has been processed. The package includes some basic descriptive functions that display the usual suspects in formats that should not agitate those accustomed to the vanilla listings that abound in the real world. prettyR is intended to assist the R beginner in producing basic stats right from the word go. No knowledge beyond that of writing an R script is required, but there is quite a bit of room to learn and innovate. Have fun and please let me know if you break it. Jim __ 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] Looking for transformation to overcome heterogeneity of variances
Thanks to all contributors for the fruitfulness of this discussion. I am speculating about a simpler solution: to use a non-parametric approach. To avoid the requirement of having normal residuals, Frank Harrell has suggested here the following non-parametric procedure: library(Design) # also requires library(Hmisc) f - lrm(y ~ a*b*c*d) f anova(f) Could someone please tell me whether that also works when there is no homoscedasticity? What are the assumptions of that method? Paul __ 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] User input from keyboard
Dear All, Can anybody tell me the syntax for User input from keyboard in R. I mean to say that if I run the program it should ask Please enter the date at the begining of the program. I am using R-2.2.1 for windows. Any help will be greatly appreciated. thanks in advance. Regards, Chiya [[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] Data frame referencing?
Dear R users, When you do: x - rnorm(10) y - rnorm(10) z - rnorm(10) a - data.frame(x,y,z) a$x [1] 1.37821893 0.21152756 -0.55453182 -2.10426048 -0.08967880 0.03712110 [7] -0.80592149 0.07413450 0.15557671 1.22165341 Why does this not work: a[a$y0.5,y] -1 Error in [-.data.frame(`*tmp*`, a$y 0.5, y, value = 1) : only 0's may be mixed with negative subscripts While this works: a[a$y0.5,2] -1 a x y z 1 1.37821893 -1.0887363 1.7340522 2 0.21152756 -0.7256467 -1.3165373 3 -0.55453182 1.000 -2.1116072 4 -2.10426048 -0.4898596 -1.5863823 5 -0.08967880 1.000 -0.9139706 6 0.03712110 1.000 -1.3004970 7 -0.80592149 -0.7004193 -0.1958059 8 0.07413450 1.000 -1.3574303 9 0.15557671 -0.3335407 -2.1991236 10 1.22165341 1.000 -0.7576708 For a complex loop I would prefer to reference the right colomn by name, not by number! Now, when the colomns change, I need to check my code to make sure that the right colomns are referenced. Suggestions much appreciated! Thanks in advance, Sander. __ 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] Data frame referencing?
When specifying a column name with [ the name must be quoted (unlike when using it with $): a[a$y 0.5, y] - 1 On 8/4/06, Sander Oom [EMAIL PROTECTED] wrote: Dear R users, When you do: x - rnorm(10) y - rnorm(10) z - rnorm(10) a - data.frame(x,y,z) a$x [1] 1.37821893 0.21152756 -0.55453182 -2.10426048 -0.08967880 0.03712110 [7] -0.80592149 0.07413450 0.15557671 1.22165341 Why does this not work: a[a$y0.5,y] -1 Error in [-.data.frame(`*tmp*`, a$y 0.5, y, value = 1) : only 0's may be mixed with negative subscripts While this works: a[a$y0.5,2] -1 a x y z 1 1.37821893 -1.0887363 1.7340522 2 0.21152756 -0.7256467 -1.3165373 3 -0.55453182 1.000 -2.1116072 4 -2.10426048 -0.4898596 -1.5863823 5 -0.08967880 1.000 -0.9139706 6 0.03712110 1.000 -1.3004970 7 -0.80592149 -0.7004193 -0.1958059 8 0.07413450 1.000 -1.3574303 9 0.15557671 -0.3335407 -2.1991236 10 1.22165341 1.000 -0.7576708 For a complex loop I would prefer to reference the right colomn by name, not by number! Now, when the colomns change, I need to check my code to make sure that the right colomns are referenced. Suggestions much appreciated! Thanks in advance, Sander. __ 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] Problem with installing R under Windows
I have been trying to install R under Windows from the source by
following the instructions in R Installation and Adminstration.
Met with a list of problems along the way.
1) Recommended packages:
I tried to do
make lnik-recommended
and
make rsync-recommended
In both cases, I got the message no rule to make target
Have I gotten the make directory wrongly?
2) Building core files:
After I did
make all recommended
in R_HOME/src/gnuwin32, the process ends with the following error
message:
make --no-print-directory -C ../modules \
OPTFLAGS='-O3 -Wall -pedantic -std=gnu99' FOPTFLAGS='-O3 -Wall' -f
Makefile.wi n gcc -shared -s -o lapack.dll lapack.def Lapack.o
dllversion.o -L../../../bin - lRlapack -lRblas -lR
c:\minGW\bin\..\lib\gcc\mingw32\3.4.5\..\..\..\..\mingw32\bin\ld.exe:
cannot fin
d -lRlapack
collect2: ld returned 1 exit status
make[5]: *** [lapack.dll] Error 1
make[4]: *** [all] Error 2
make[3]: *** [all] Error 1
make[2]: *** [rmodules] Error 2
make[1]: *** [rbuild] Error 2
make: *** [all] Error 2
What should I be doing here?
3) Building manuals
After I did
make manuals
in R_HOME/src/gnuwin32, the process ends with the error message
! LaTeX Error: File `datasets-pkg.tex' not found.
Type X to quit or RETURN to proceed,
or enter new name. (Default extension: tex)
Enter file name:
! Emergency stop.
read *
l.77 \input{datasets-pkg.tex}
! == Fatal error occurred, the output PDF file is not finished!
Transcript written on refman.log.
make[1]: *** [refman.pdf] Error 1
make: *** [manuals] Error 2
What should I be doing here?
Can anyone help?
Thanks.
Rgds
Chong Hui
[[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] between-within anova: aov and lme
Well nobody answered :-( Nobody on R-help is doing anovas I guess -- I don't blame them! (It's just for aggies.) In the absence of any response and for no good reason I am doing: fitn1 - aov(amplitude ~ stereo*site*stimulus + Error(subject), stereon1) This is Bill Venables's way. And when the data are unbalanced I am doing: lme(amplitude ~ site+stimulus+stereo*stimulus, random=~1|subject, method=ML, stereon1) And I have no clue why. Every discussion of between-within ANOVA I have read (practical or mathematical) is either vacuous or opaque... Cheers Bill I have 2 questions on ANOVA with 1 between subjects factor and 2 within factors. 1. I am confused on how to do the analysis with aov because I have seen two examples on the web with different solutions. a) Jon Baron (http://www.psych.upenn.edu/~baron/rpsych/rpsych.html) does 6.8.5 Example 5: Stevens pp. 468 - 474 (one between, two within) between: gp within: drug, dose aov(effect ~ gp * drug * dose + Error(subj/(dose*drug)), data=Ela.uni) b) Bill Venables answered a question on R help as follows. - factor A between subjects - factors B*C within subjects. aov(response ~ A*B*C + Error(subject), Kirk) An alternative formula would be response ~ A/(B*C) + Error(subject), which would only change things by grouping together some of the sums of squares. --- SO: which should I do aov(response ~ A*B*C + Error(subject), Kirk) aov(response ~ A/(B*C) + Error(subject), Kirk) aov(response ~ A*B*C + Error(subject/(B*C)), Kirk) 2. How would I do the analysis in lme()? Something like lme(response~A*B*C,random=~1|subject/(B*C))??? Thanks very much for any help! Bill Simpson __ 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] Data frame referencing?
you need to use quotes, i.e., a[a$y 0.5, y] - 1 you can also use a$y[a$y 0.5] - 1 I hope it helps. Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/(0)16/336899 Fax: +32/(0)16/337015 Web: http://med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: Sander Oom [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Sent: Friday, August 04, 2006 1:48 PM Subject: [R] Data frame referencing? Dear R users, When you do: x - rnorm(10) y - rnorm(10) z - rnorm(10) a - data.frame(x,y,z) a$x [1] 1.37821893 0.21152756 -0.55453182 -2.10426048 -0.08967880 0.03712110 [7] -0.80592149 0.07413450 0.15557671 1.22165341 Why does this not work: a[a$y0.5,y] -1 Error in [-.data.frame(`*tmp*`, a$y 0.5, y, value = 1) : only 0's may be mixed with negative subscripts While this works: a[a$y0.5,2] -1 a x y z 1 1.37821893 -1.0887363 1.7340522 2 0.21152756 -0.7256467 -1.3165373 3 -0.55453182 1.000 -2.1116072 4 -2.10426048 -0.4898596 -1.5863823 5 -0.08967880 1.000 -0.9139706 6 0.03712110 1.000 -1.3004970 7 -0.80592149 -0.7004193 -0.1958059 8 0.07413450 1.000 -1.3574303 9 0.15557671 -0.3335407 -2.1991236 10 1.22165341 1.000 -0.7576708 For a complex loop I would prefer to reference the right colomn by name, not by number! Now, when the colomns change, I need to check my code to make sure that the right colomns are referenced. Suggestions much appreciated! Thanks in advance, Sander. __ 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. 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] User input from keyboard
Hi cat(\n,Enter x,\n) # prompt y-scan(n=1) prompts for user imput and scans 1 line from console. HTH Petr On 4 Aug 2006 at 17:06, chiya sharma wrote: Date sent: Fri, 4 Aug 2006 17:06:38 +0530 From: chiya sharma [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Subject:[R] User input from keyboard Dear All, Can anybody tell me the syntax for User input from keyboard in R. I mean to say that if I run the program it should ask Please enter the date at the begining of the program. I am using R-2.2.1 for windows. Any help will be greatly appreciated. thanks in advance. Regards, Chiya [[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. 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] Data frame referencing?
Dear Gabor and Dimitris, Simple, once you know! It is these little exceptions on the R notation that get me stuck. Now I am on the loose again! Thanks, Sander. Dimitris Rizopoulos wrote: you need to use quotes, i.e., a[a$y 0.5, y] - 1 you can also use a$y[a$y 0.5] - 1 I hope it helps. Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/(0)16/336899 Fax: +32/(0)16/337015 Web: http://med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: Sander Oom [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Sent: Friday, August 04, 2006 1:48 PM Subject: [R] Data frame referencing? Dear R users, When you do: x - rnorm(10) y - rnorm(10) z - rnorm(10) a - data.frame(x,y,z) a$x [1] 1.37821893 0.21152756 -0.55453182 -2.10426048 -0.08967880 0.03712110 [7] -0.80592149 0.07413450 0.15557671 1.22165341 Why does this not work: a[a$y0.5,y] -1 Error in [-.data.frame(`*tmp*`, a$y 0.5, y, value = 1) : only 0's may be mixed with negative subscripts While this works: a[a$y0.5,2] -1 a x y z 1 1.37821893 -1.0887363 1.7340522 2 0.21152756 -0.7256467 -1.3165373 3 -0.55453182 1.000 -2.1116072 4 -2.10426048 -0.4898596 -1.5863823 5 -0.08967880 1.000 -0.9139706 6 0.03712110 1.000 -1.3004970 7 -0.80592149 -0.7004193 -0.1958059 8 0.07413450 1.000 -1.3574303 9 0.15557671 -0.3335407 -2.1991236 10 1.22165341 1.000 -0.7576708 For a complex loop I would prefer to reference the right colomn by name, not by number! Now, when the colomns change, I need to check my code to make sure that the right colomns are referenced. Suggestions much appreciated! Thanks in advance, Sander. __ 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. 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. __ 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 special token \\ from R to latex
Dear helpeRs,
I would like to specify a newline command in R and pass it to latex
via Sweave such that it corresponds to latex' \\ command. But that
doesn't seem to be possible. If I Sweave the \n character, it just
makes a new line in latex but not the \\ command.
Is there a way such that the following code would result in latex in
blablabla \\ blablabla on different line
echo=FALSE=
string - blablabla \\ blablabla on different line
@
\Sexpr{string}
Thanks for the help,
Jan
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] User input from keyboard
On Fri, 4 Aug 2006, Petr Pikal wrote: Hi cat(\n,Enter x,\n) # prompt y-scan(n=1) prompts for user imput and scans 1 line from console. (that scans one number: use readLines(n=1) to get a string). But readline() is probably easier. HTH Petr On 4 Aug 2006 at 17:06, chiya sharma wrote: Date sent:Fri, 4 Aug 2006 17:06:38 +0530 From: chiya sharma [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Subject: [R] User input from keyboard Dear All, Can anybody tell me the syntax for User input from keyboard in R. I mean to say that if I run the program it should ask Please enter the date at the begining of the program. I am using R-2.2.1 for windows. -- 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] Sweave special token \\ from R to latex
Jan Wijffels wrote:
Dear helpeRs,
I would like to specify a newline command in R and pass it to latex
via Sweave such that it corresponds to latex' \\ command. But that
doesn't seem to be possible. If I Sweave the \n character, it just
makes a new line in latex but not the \\ command.
Is there a way such that the following code would result in latex in
blablabla \\ blablabla on different line
echo=FALSE=
string - blablabla \\ blablabla on different line
@
\Sexpr{string}
For reasons that I am unable to account for, you get the desired string
in your .tex file if you do
echo=FALSE=
string - blablabla blablabla on different line
@
\Sexpr{print(string)}
resulting in the .tex file as
blablabla \\ blablabla on different line
It would seem that using echo=FALSE,results=tex
would be a good solution.
Markus
Thanks for the help,
Jan
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.
--
Markus Jantti
Abo Akademi University
[EMAIL PROTECTED]
http://www.iki.fi/~mjantti
###
This message has been scanned by F-Secure Anti-Virus for Mic...{{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] plotting picture data
Just a suggestion. It seems like each square can be denoted by x and y coordinates. Then you essentially have a two dimensional histogram/density that you need to plot. You can use the lattice functions cloud/wireframe. You can also go for a heat map/contour plot, the lattice functions for that will be levelplot/contourplot. In case the number of squares are small, you might prefer a two-dimensional histogram, cloud in lattice has an option to plot the point as histogram. Ritwik. On 8/4/06, Gichangi, Anthony [EMAIL PROTECTED] wrote: Hi R users I have a dataset which represents points that are market by patients as the source of pain. Basically the patients indicates by a cross on a chest pictures where he/she thinks is the source of pain. The data was then digitalized by divinding the chest into small squares and each square was give value 1 if it was the center 2 if it was touched by the markings and 3 if it was not touched. I would like to plot this data on the chest like graph showing the intesities of different points and later stratify the grouping variables to see the difference. Has anybody got an idea how I can go around this ? Help is highly appreciated. Regards Anthony Gichangi, M. sc. Department of Statistics. JB. Winsløvej 9B, DK 5000 Odense C. Tel: 00 45 6550 3379 Mobile: 00 45 61105805 __ 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. -- Ritwik Sinha Graduate Student Epidemiology and Biostatistics Case Western Reserve University http://darwin.cwru.edu/~rsinha [[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] plotting picture data
Hi seems to me that it can be done by image. See ?image. Just as illustration. mat-matrix(sample(c(1,rep(2,10), rep(3,50)), 1000, replace=T), 100,100) for(i in 1:6) mat[i,c(1:(50-5*i),(50+5*i):100)]-NA for(i in 14:9) mat[i,c(1:(50-5*(15-i)),(50+5*(15-i)):100)]-NA image(1:100,1:100,mat) HTH Petr On 4 Aug 2006 at 15:12, Gichangi, Anthony wrote: From: Gichangi, Anthony [EMAIL PROTECTED] To: R-help r-help@stat.math.ethz.ch Date sent: Fri, 4 Aug 2006 15:12:36 +0200 Subject:[R] plotting picture data Hi R users I have a dataset which represents points that are market by patients as the source of pain. Basically the patients indicates by a cross on a chest pictures where he/she thinks is the source of pain. The data was then digitalized by divinding the chest into small squares and each square was give value 1 if it was the center 2 if it was touched by the markings and 3 if it was not touched. I would like to plot this data on the chest like graph showing the intesities of different points and later stratify the grouping variables to see the difference. Has anybody got an idea how I can go around this ? Help is highly appreciated. Regards Anthony Gichangi, M. sc. Department of Statistics. JB. Winslřvej 9B, DK 5000 Odense C. Tel: 00 45 6550 3379 Mobile: 00 45 61105805 __ 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] read.spss and umlaut
On Fri, 4 Aug 2006, Thomas Kuster wrote: If I change the translatable characters (overwrite the 0 between :#@'= and ~000 with ÄÖÜäöü). I can read in the file an every ÄÖÜäöü ist a withspace: Ok, that's the problem then. The file format says that the umlauts are unreadable and R is believing the file format. I will look at adding an option to specify an encoding and ignore the translation table, but not very urgently. -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] question
Hi Mr Plate, I have a little question How to convert a rowname vector of numbers into a real column of the matrix, My problem is that I applied a rowsum function on a matrix. Then I get a matrix in which the names of the columns are the values of the group (numbers) Now I need to make calculation on the groups row. How to convert this vector of (rownames) into a real column in the matrix ? thanks by advance Jessica Gervais _ www.windowslivemessenger.fr __ 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] expression() - Superscript in y-axis, keeping line break in string
I've tried several different ways to accomplish this, but as yet to no avail. My y-axis for a plot has a rather long label, and thus I have been using /n to break it into two lines. However, to make it technically correct for publication, I also need to use superscript in the label. For example: par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1) plot(1:10, ylab=14C-glyphosate line1\n line2) will provide the text in two lines as I would like it. However, I am trying to keep those same line breaks when using expression() to get my superscript number. This will not work, as it aligns the 14C section with the bottom line of the expression making little sense to the reader. par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1) plot(1:10, ylab=expression( ^14*C*-glyphosate line1\n line2)) Is there a way to align the 14C portion of the expression with the top line of the string rather than the bottom line? Any suggestions are greatly appreciated. Andrew -- Andrew Kniss Assistant Research Scientist University of Wyoming Department of Plant Sciences [EMAIL PROTECTED] Office: (307) 766-3949 Fax:(307) 766-5549 __ 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] expression() - Superscript in y-axis, keeping line break in string
On Fri, 2006-08-04 at 09:47 -0600, Andrew Kniss wrote: I've tried several different ways to accomplish this, but as yet to no avail. My y-axis for a plot has a rather long label, and thus I have been using /n to break it into two lines. However, to make it technically correct for publication, I also need to use superscript in the label. For example: par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1) plot(1:10, ylab=14C-glyphosate line1\n line2) will provide the text in two lines as I would like it. However, I am trying to keep those same line breaks when using expression() to get my superscript number. This will not work, as it aligns the 14C section with the bottom line of the expression making little sense to the reader. par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1) plot(1:10, ylab=expression( ^14*C*-glyphosate line1\n line2)) Is there a way to align the 14C portion of the expression with the top line of the string rather than the bottom line? Any suggestions are greatly appreciated. Andrew plotmath, as has been covered many times previously, does not support multi-line expressions. A note should probably be added to ?plotmath on this. Thus, you need to create each line in the label separately: par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1) plot(1:10, ylab = ) # Now use mtext() to place each line of the y axis label mtext(2, text = expression( ^14*C*-glyphosate line1), line = 3) mtext(2, text = line2, line = 2) See ?mtext for more information. 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] expression() - Superscript in y-axis, keeping line break in string
Use atop: plot(1, main = expression(atop( ^14*C*-glyphosate line, line2))) On 8/4/06, Andrew Kniss [EMAIL PROTECTED] wrote: I've tried several different ways to accomplish this, but as yet to no avail. My y-axis for a plot has a rather long label, and thus I have been using /n to break it into two lines. However, to make it technically correct for publication, I also need to use superscript in the label. For example: par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1) plot(1:10, ylab=14C-glyphosate line1\n line2) will provide the text in two lines as I would like it. However, I am trying to keep those same line breaks when using expression() to get my superscript number. This will not work, as it aligns the 14C section with the bottom line of the expression making little sense to the reader. par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1) plot(1:10, ylab=expression( ^14*C*-glyphosate line1\n line2)) Is there a way to align the 14C portion of the expression with the top line of the string rather than the bottom line? Any suggestions are greatly appreciated. Andrew -- Andrew Kniss Assistant Research Scientist University of Wyoming Department of Plant Sciences [EMAIL PROTECTED] Office: (307) 766-3949 Fax:(307) 766-5549 __ 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] expression() - Superscript in y-axis, keeping line break in string
Sorry, you wanted a ylab=, not a main=. Try using xyplot in lattice: library(lattice) xyplot(1~1, ylab = expression(atop(phantom(0)^14*C*-glyphosate line, line2))) On 8/4/06, Gabor Grothendieck [EMAIL PROTECTED] wrote: Use atop: plot(1, main = expression(atop( ^14*C*-glyphosate line, line2))) On 8/4/06, Andrew Kniss [EMAIL PROTECTED] wrote: I've tried several different ways to accomplish this, but as yet to no avail. My y-axis for a plot has a rather long label, and thus I have been using /n to break it into two lines. However, to make it technically correct for publication, I also need to use superscript in the label. For example: par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1) plot(1:10, ylab=14C-glyphosate line1\n line2) will provide the text in two lines as I would like it. However, I am trying to keep those same line breaks when using expression() to get my superscript number. This will not work, as it aligns the 14C section with the bottom line of the expression making little sense to the reader. par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1) plot(1:10, ylab=expression( ^14*C*-glyphosate line1\n line2)) Is there a way to align the 14C portion of the expression with the top line of the string rather than the bottom line? Any suggestions are greatly appreciated. Andrew -- Andrew Kniss Assistant Research Scientist University of Wyoming Department of Plant Sciences [EMAIL PROTECTED] Office: (307) 766-3949 Fax:(307) 766-5549 __ 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] meta characters in file path
Thanks. I tried them, it works for most of those characters except * and ?. Does regular expression work in file names in windows? e.g. I have a machine-generated file named 021706 matrix#1479 @50.csv, of which 1479 is kinda random. Will I be able to match 1479 with some sort of wild card chars? Thanks Johnny -Original Message- From: Tony Plate [mailto:[EMAIL PROTECTED] Sent: Thursday, August 03, 2006 3:42 PM To: Li,Qinghong,ST.LOUIS,Molecular Biology Cc: r-help@stat.math.ethz.ch Subject: Re: [R] meta characters in file path What is the problem you are having? Seems to work fine for me running under Windows2000: write.table(data.frame(a=1:3,b=4:6), file=@# x.csv, sep=,) read.csv(file=@# x.csv) a b 1 1 4 2 2 5 3 3 6 sessionInfo() Version 2.3.1 (2006-06-01) i386-pc-mingw32 attached base packages: [1] methods stats graphics grDevices utils datasets [7] base other attached packages: XML 0.99-8 Li,Qinghong,ST.LOUIS,Molecular Biology wrote: Hi, I need to read in some files. The file names contain come meta characters such as @, #, and white spaces etc, In read.csv, file= option, is there any way that one can make the function to recognize a file path with those characters? Thanks Johnny [[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] expression() - Superscript in y-axis, keeping line break in string
Actually Gabor, using your solution with 'atop', which I had not considered, it will work with base graphics: par(oma = c(0, 0, 2, 0), mar = c(5, 6, 0.25, 2), lheight = 1) plot(1:10, ylab = expression(atop( ^14*C*-glyphosate line1, line2))) HTH, Marc On Fri, 2006-08-04 at 12:09 -0400, Gabor Grothendieck wrote: Sorry, you wanted a ylab=, not a main=. Try using xyplot in lattice: library(lattice) xyplot(1~1, ylab = expression(atop(phantom(0)^14*C*-glyphosate line, line2))) On 8/4/06, Gabor Grothendieck [EMAIL PROTECTED] wrote: Use atop: plot(1, main = expression(atop( ^14*C*-glyphosate line, line2))) On 8/4/06, Andrew Kniss [EMAIL PROTECTED] wrote: I've tried several different ways to accomplish this, but as yet to no avail. My y-axis for a plot has a rather long label, and thus I have been using /n to break it into two lines. However, to make it technically correct for publication, I also need to use superscript in the label. For example: par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1) plot(1:10, ylab=14C-glyphosate line1\n line2) will provide the text in two lines as I would like it. However, I am trying to keep those same line breaks when using expression() to get my superscript number. This will not work, as it aligns the 14C section with the bottom line of the expression making little sense to the reader. par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1) plot(1:10, ylab=expression( ^14*C*-glyphosate line1\n line2)) Is there a way to align the 14C portion of the expression with the top line of the string rather than the bottom line? Any suggestions are greatly appreciated. Andrew __ 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] Simulate an Overdispersed(extra-variance poisson process)?
Is there a function in R comparable to rpois that can simulate random variables from an overdispersed poisson distribution? If there is not a function any ideas/references on how to program one? thanks, Spencer Jones Graduate Student, NLM Fellow Dept. Biomedical Informatics University of Utah [[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] Simulate an Overdispersed(extra-variance poisson process)?
On Fri, 2006-08-04 at 10:33 -0600, Spencer Jones wrote: Is there a function in R comparable to rpois that can simulate random variables from an overdispersed poisson distribution? If there is not a function any ideas/references on how to program one? Take a look at ?rnbinom or library(MASS) ?rnegbin 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.
[R] Error when loading odesolve
Dear list, I installed odesolve package (0.5-15) in R 2.3.1 in a Solaris server (Generic_118558-11 sun4u sparc SUNW,Sun-Blade-1000). The installing progress completed without errors, though several warnings like Warning: Option -fPIC passed to ld, if ld is invoked, ignored otherwise were outputed. However, when loading the odesolve package by library(odesolve), following error messages pop out: library(odesolve) Error in dyn.load(x, as.logical(local), as.logical(now)) : unable to load shared library '/project/scratch/ligroup/R1/lib/R/library/odesolve/libs/odesolve.so': ld.so.1: R: fatal: relocation error: file /project/scratch/ligroup/R1/lib/R/library/odesolve/libs/odesolve.so: symbol __f90_ssfw: referenced symbol not found Error: package/namespace load failed for 'odesolve' Could any one tell me how to fix this problem? Thanks very much. Wuming __ 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] Sampling from a Matrix
On Fri, 2006-08-04 at 12:46 -0400, Daniel Gerlanc wrote:
Hello all,
Consider the following problem:
There is a matrix of probabilities:
set.seed(1)
probs - array(abs(rnorm(25, sd = 0.33)), dim = c(5,5), dimnames =
list(1:5, letters[1:5]))
probs
a b c de
1 0.21 0.27 0.50 0.0148 0.303
2 0.06 0.16 0.13 0.0053 0.258
3 0.28 0.24 0.21 0.3115 0.025
4 0.53 0.19 0.73 0.2710 0.656
5 0.11 0.10 0.37 0.1960 0.205
I want to sample 3 values from each row.
One way to do this follows:
index - 1:ncol(probs)
for(i in 1:nrow(probs)){
## gets the indexes of the values chosen
sample(index, size = 3, replace = TRUE, prob = probs[i, ])
}
Is there a another way to do this?
Thanks!
t(apply(probs, 1, function(x) sample(x, 3)))
[,1] [,2] [,3]
1 0.210 0.5000 0.0148
2 0.258 0.0053 0.1300
3 0.025 0.2800 0.3115
4 0.190 0.5300 0.2710
5 0.196 0.1000 0.1100
See ?apply and ?t
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.
[R] GAM 2D-plotting
Hi, When I fit a GAM model (using mgvc) with overlapping terms, such as gam(y~s(x,z)+s(z,w)) and afterwards I pretend to plot the component smooth functions that make it up using plot.gam, I achieve a couple of 2D plots. My question is: What's the meaning of those 2D plots in terms of y? Regards, Nixon __ 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] training svm's with probability flag
Hi- I'm seeing some weirdness with svm and tune.svm that I can't figure out- was wondering if anyone else has seen this? Perhaps I'm failing to make something the expected class? Below is my repro case, though it *sometimes* doesn't repro. I'm using R2.3.1 on WindowsXP. I was also seeing it happen with R2.1.1 and have seen it on 2 different machines. data(iris) attach(iris) library(e1071) train- iris[c(1:30,50:80,100:130),] test- iris[-c(1:30,50:80,100:130),] y.train- train$Species y.test- test$Species obj- tune.svm(train[,-5], y.train, gamma = 2^(-1:1), cost = 2^(2:4), probability=T) my.svm- obj$best.model pred1- predict(my.svm, test[,-5]) pred2- predict(my.svm, test[,-5], probability=T) table(pred1, y.test) table(pred2, y.test) When I do this, the two different tables often come out different, as below: table(pred1, y.test) y.test pred1setosa versicolor virginica setosa 19 0 0 versicolor 0 18 1 virginica 0 119 table(pred2, y.test) y.test pred2setosa versicolor virginica setosa 18 0 0 versicolor 1 18 1 virginica 0 119 I'm not sure 1. why the results would differ based on whether I choose to calculate the probabilities, and 2. which one to trust?? Anyone come across this before, or have any ideas? thanks, jessie [[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] why does lm() not allow for negative weights?
Dear List, Why do commonly used estimator functions (such as lm(), glm(), etc.) not allow negative case weights? I suspect that there is a good reason for this. Yet, I can see reasonable cases when one wants to use negative case weights. Take lm() for example: ### n - 20 Y - rnorm(n) X - cbind(rep(1,n),runif(n),rnorm(n)) Weights - rnorm(n) # Includes Pos and Neg Weights Weights # Now do Weighted LS and get beta coeffs: b - solve(t(X)%*%diag(Weights)%*%X) %*% t(X) %*% diag(Weights)%*%Y b # This seems like a valid model, but when I try lm(Y ~ X[,2:3],weights=Weights) # I get: missing or negative weights not allowed ### What is the rationale for not allowing negative weights? I ask this, because I am currently trying to implement a (two stage) estimator into R that involves negative case weights. Weights are generated in the first stage, so it would be nice if I could use canned functions such as lm(,weights=Weights) in the second stage. Thank you for your help. Best, Jens __ 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] meta characters in file path
On Fri, 4 Aug 2006, Li,Qinghong,ST.LOUIS,Molecular Biology wrote: Thanks. I tried them, it works for most of those characters except * and ?. Those are not valid characters in Windows file paths (/ * : ? \ | are invalid in file or dir names). Does regular expression work in file names in windows? No, and I think you may mean wildcards (which is what work on the command line). e.g. I have a machine-generated file named 021706 matrix#1479 @50.csv, of which 1479 is kinda random. Will I be able to match 1479 with some sort of wild card chars? Yes, use dir(), with regexp pattern patching to find the name(s) you want. glob2rx() might be useful here. Thanks Johnny -Original Message- From: Tony Plate [mailto:[EMAIL PROTECTED] Sent: Thursday, August 03, 2006 3:42 PM To: Li,Qinghong,ST.LOUIS,Molecular Biology Cc: r-help@stat.math.ethz.ch Subject: Re: [R] meta characters in file path What is the problem you are having? Seems to work fine for me running under Windows2000: write.table(data.frame(a=1:3,b=4:6), file=@# x.csv, sep=,) read.csv(file=@# x.csv) a b 1 1 4 2 2 5 3 3 6 sessionInfo() Version 2.3.1 (2006-06-01) i386-pc-mingw32 attached base packages: [1] methods stats graphics grDevices utils datasets [7] base other attached packages: XML 0.99-8 Li,Qinghong,ST.LOUIS,Molecular Biology wrote: Hi, I need to read in some files. The file names contain come meta characters such as @, #, and white spaces etc, In read.csv, file= option, is there any way that one can make the function to recognize a file path with those characters? Thanks Johnny [[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. -- 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] Error when loading odesolve
Looking at odesolve, src/Makevars is PKG_LIBS=$(BLAS_LIBS) Now, the documentation says that if you have $(BLAS_LIBS) you must also have $(FLIBS), so please change this to PKG_LIBS=$(BLAS_LIBS) $(FLIBS) and take this up with the package maintainer (which is what the posting guide asked you to do in the first place). On Sat, 5 Aug 2006, Wuming Gong wrote: Dear list, I installed odesolve package (0.5-15) in R 2.3.1 in a Solaris server (Generic_118558-11 sun4u sparc SUNW,Sun-Blade-1000). The installing progress completed without errors, though several warnings like Warning: Option -fPIC passed to ld, if ld is invoked, ignored otherwise were outputed. However, when loading the odesolve package by library(odesolve), following error messages pop out: library(odesolve) Error in dyn.load(x, as.logical(local), as.logical(now)) : unable to load shared library '/project/scratch/ligroup/R1/lib/R/library/odesolve/libs/odesolve.so': ld.so.1: R: fatal: relocation error: file /project/scratch/ligroup/R1/lib/R/library/odesolve/libs/odesolve.so: symbol __f90_ssfw: referenced symbol not found Error: package/namespace load failed for 'odesolve' Could any one tell me how to fix this problem? Thanks very much. Wuming __ 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] why does lm() not allow for negative weights?
On 8/4/2006 1:26 PM, Jens Hainmueller wrote: Dear List, Why do commonly used estimator functions (such as lm(), glm(), etc.) not allow negative case weights? Residual sums of squares (or deviances) could be negative with negative case weights. This doesn't seem like a good thing: would you really want the fit to be far from those points? I suspect that there is a good reason for this. Yet, I can see reasonable cases when one wants to use negative case weights. Take lm() for example: ### n - 20 Y - rnorm(n) X - cbind(rep(1,n),runif(n),rnorm(n)) Weights - rnorm(n) # Includes Pos and Neg Weights Weights # Now do Weighted LS and get beta coeffs: b - solve(t(X)%*%diag(Weights)%*%X) %*% t(X) %*% diag(Weights)%*%Y That formula does not necessarily give least squares estimates in the case where weights might be negative. For example, with a single observation y, a single parameter mu, design matrix X = 1, and weight -1, that formula becomes b - y, but that is the worst possible estimator in a least squares sense. The residual sum of squares can be made arbitrarily large and negative by setting b to a large value. Duncan Murdoch b # This seems like a valid model, but when I try lm(Y ~ X[,2:3],weights=Weights) # I get: missing or negative weights not allowed ### What is the rationale for not allowing negative weights? I ask this, because I am currently trying to implement a (two stage) estimator into R that involves negative case weights. Weights are generated in the first stage, so it would be nice if I could use canned functions such as lm(,weights=Weights) in the second stage. Thank you for your help. Best, Jens __ 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] why does lm() not allow for negative weights?
Thanks Duncan Murdoch, Why do commonly used estimator functions (such as lm(), glm(), etc.) not allow negative case weights? Residual sums of squares (or deviances) could be negative with negative case weights. This doesn't seem like a good thing: would you really want the fit to be far from those points? Yes, this is actually what I want for this particular estimator. But I can see now why this generally doesn't seem like a a good idea. Best, Jens -Ursprüngliche Nachricht- Von: Duncan Murdoch [mailto:[EMAIL PROTECTED] Gesendet: Friday, August 04, 2006 7:36 PM An: Jens Hainmueller Cc: r-help@stat.math.ethz.ch Betreff: Re: [R] why does lm() not allow for negative weights? On 8/4/2006 1:26 PM, Jens Hainmueller wrote: Dear List, I suspect that there is a good reason for this. Yet, I can see reasonable cases when one wants to use negative case weights. Take lm() for example: ### n - 20 Y - rnorm(n) X - cbind(rep(1,n),runif(n),rnorm(n)) Weights - rnorm(n) # Includes Pos and Neg Weights Weights # Now do Weighted LS and get beta coeffs: b - solve(t(X)%*%diag(Weights)%*%X) %*% t(X) %*% diag(Weights)%*%Y That formula does not necessarily give least squares estimates in the case where weights might be negative. For example, with a single observation y, a single parameter mu, design matrix X = 1, and weight -1, that formula becomes b - y, but that is the worst possible estimator in a least squares sense. The residual sum of squares can be made arbitrarily large and negative by setting b to a large value. Duncan Murdoch b # This seems like a valid model, but when I try lm(Y ~ X[,2:3],weights=Weights) # I get: missing or negative weights not allowed ### What is the rationale for not allowing negative weights? I ask this, because I am currently trying to implement a (two stage) estimator into R that involves negative case weights. Weights are generated in the first stage, so it would be nice if I could use canned functions such as lm(,weights=Weights) in the second stage. Thank you for your help. Best, Jens __ 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] why does lm() not allow for negative weights?
On 4 August 2006 at 19:26, Jens Hainmueller wrote: | Why do commonly used estimator functions (such as lm(), glm(), etc.) not | allow negative case weights? I suspect that there is a good reason for this. That came up on r-sig-finance a little while ago. As usual, Gabor won the contest for most precise and concise answer with: [..] At any rate, note that if the weights can be negative then the sum of squares to be optimized is no longer a convex function of the coefficients so we really don't have a conventional least squares model and uniqueness and existence have possibly different answers. See the r-sig-finance archives for April 2006 and a thread entitled 'negative weights'. Hth, Dirk -- Hell, there are no rules here - we're trying to accomplish something. -- Thomas A. Edison __ 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] prettyR arrives
On Fri, 2006-08-04 at 19:37 -0400, Jim Lemon wrote: Hi all, I have finally gotten the prettyR package going (many thanks to Kurt Hornik for his patience). prettyR is a set of functions that allows the user to produce HTML output from R scripts. Given an R script that runs properly, an HTML listing complete with embedded graphics can be produced simply by passing the script to the core function htmlize (Phillipe Grosjean has not only offered great suggestions, but provided a fancier function named R2html). It is even possible to have the output magically appear in your friendly local HTML browser when the script has been processed. The package includes some basic descriptive functions that display the usual suspects in formats that should not agitate those accustomed to the vanilla listings that abound in the real world. prettyR is intended to assist the R beginner in producing basic stats right from the word go. No knowledge beyond that of writing an R script is required, but there is quite a bit of room to learn and innovate. Have fun and please let me know if you break it. Jim __ 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. Thanks but I could not get R2html in prettyR to work: R2html(Rfile=/ophth/cornea/R/lme_4.R, + HTMLfile=/ophth/cornea/Reports/lme_4.html) Error in CreateIndexFile(HTMLfile, basenavfile, baselistfile, title) : unused argument(s) ( ...) lme_3.r has the R script and lme_3.html is the html file I'd like to create. The help file for R2html does not give an example. args(R2html) function (Rfile, HTMLfile, echo = TRUE, split = FALSE, browse = TRUE, title = R listing, bgcolor = #dd, ...) What am I doing wrong? I can source the script file. args(CreateIndexFile) function (HTMLbase, HTMLdir, title = R listing) Is there a problem in R2html's call to CreateIndexFile? The arguments don't seem to match. Rick B. __ 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] ggplot facet label font size
With ggplot its possible to do this too but in that case it seems
necessary to recurse through the grobs. Here we look for
grobs that have a label component which contains a colon
and grid.edit those changing the value of cex. Note that
getNames() give a single grob named pretty and we start
from that:
# run code
library(ggplot)
library(grid)
p - ggplot(tips, sex ~ smoker, aesthetics=list(x=tip/total_bill))
gghistogram(p)
recurse - function(x) {
if (!is.null(x$label) regexpr(:, x$label) 0)
grid.edit(x$name, gp = gpar(cex = 0.7))
for (ch in x$children) recurse(ch)
}
recurse(grid.get(pretty))
On 8/3/06, Gabor Grothendieck [EMAIL PROTECTED] wrote:
If you are willing to use grid then you could create only the sex
factor in the left strips since its already in the desired position
but when displaying it output a factor.level, i.e. label of A.
(my.strip.left is modified from the prior post to do that.)
Then after the plot is drawn, looping through all grobs looking for
those with a label component of A producing a list of grob names,
strip.left.names. We then mapply the real factor levels with
those grobs editing them in reset.levels(), defined below.
(I have used the fact, empirically determined that the stripts
are produced in order of the factor levels.)
Everything is the same as the last post except my.strip.left
which has been modified and everything which comes after the
call to histogram.
Although this seems to work, maybe Deepayan or Paul can think of
something slicker.
library(ggplot) # data resides here
library(lattice)
library(grid)
my.strip - function(which.given, which.panel, ...)
if (which.given == 1 which.panel[2] == 2)
strip.default(which.given, which.panel, ...)
my.strip.left - function(which.given, which.panel, ...,
factor.levels, horizontal)
if (which.given == 1 which.panel[1] == 1)
strip.default(which.given, which.panel, factor.levels = LETTERS,
horizontal = FALSE, ...)
histogram(~ tip/total_bill | sex + smoker, tips, strip = my.strip,
strip.left = my.strip.left, par.settings = list(add.text =
list(cex = 0.7)))
is.strip.left - function(name) identical(grid.get(name)$label, A)
strip.left.names - getNames()[sapply(getNames(), is.strip.left)]
reset.levels - function(nam, lev) grid.edit(nam, label = lev)
mapply(reset.levels , strip.left.names, levels(tips$smoker))
On 8/3/06, Walker, Sam [EMAIL PROTECTED] wrote:
This works OK, but there is some extra spacing between the panels, the
top axis and the strip on the top, and the left labels and panel.
How can I remove these extra spaces?
I've tried changing various layout.widths settings with no luck. It
seems the spaces are calculated based on the number of conditioning
variables, in this case 2 (sex+smoker).
Thanks in advance...
-Sam
-Original Message-
From: Gabor Grothendieck [mailto:[EMAIL PROTECTED]
Sent: Wednesday, August 02, 2006 6:04 PM
To: Walker, Sam
Cc: r-help@stat.math.ethz.ch
Subject: Re: [R] ggplot facet label font size
On 8/2/06, Walker, Sam [EMAIL PROTECTED] wrote:
How do I change the font size in the facet labels along the edges of
the
plot?
For example (from the ggplot help file):
p-ggplot(tips, sex ~ smoker, aesthetics=list(x=tip/total_bill))
gghistogram(p)
In this plot, the facet labels are smoker: No, smoker: Yes, sex:
Female, sex: Male. What command can I use to reduce the font size
of
these labels?
In lattice terminology, cex is used to scale these strip labels. But
I
couldn't find the equivalent in ggplot.
The reason I'm asking is I have a 9x7 array of plots which I've been
plotting with lattice. I wanted to use ggplot because I like having
the
labels on the edge of the plots
Note that lattice can do that by using custom strip functions:
library(ggplot) # data resides here
library(lattice)
my.strip - function(which.given, which.panel, ...)
if (which.given == 1 which.panel[2] == 2)
strip.default(which.given, which.panel, ...)
my.strip.left - function(which.given, which.panel, ..., horizontal)
if (which.given == 2 which.panel[1] == 1)
strip.default(which.given, which.panel, horizontal = FALSE, ...)
histogram(~ tip/total_bill | sex + smoker, tips, strip = my.strip,
strip.left = my.strip.left, par.settings = list(add.text =
list(cex = 0.7)))
__
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,
Re: [R] expression() - Superscript in y-axis, keeping line break in string
On Fri, 4 Aug 2006, Marc Schwartz (via MN) wrote:
On Fri, 2006-08-04 at 09:47 -0600, Andrew Kniss wrote:
I've tried several different ways to accomplish this, but as yet to no
avail. My y-axis for a plot has a rather long label, and thus I have
been using /n to break it into two lines. However, to make it
technically correct for publication, I also need to use superscript in
the label. For example:
par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1)
plot(1:10,
ylab=14C-glyphosate line1\n line2)
will provide the text in two lines as I would like it. However, I am
trying to keep those same line breaks when using expression() to get my
superscript number. This will not work, as it aligns the 14C section
with the bottom line of the expression making little sense to the
reader.
par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1)
plot(1:10,
ylab=expression( ^14*C*-glyphosate line1\n line2))
Is there a way to align the 14C portion of the expression with the top
line of the string rather than the bottom line? Any suggestions are
greatly appreciated.
Andrew
plotmath, as has been covered many times previously, does not support
multi-line expressions. A note should probably be added to ?plotmath on
this.
I've added a note. I think what is exact is that control chars are not
interpreted ('expresssion' is an overloaded work in this context).
Thanks for the nudge (and please do continue to make such remarks).
Brian
Thus, you need to create each line in the label separately:
par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1)
plot(1:10, ylab = )
# Now use mtext() to place each line of the y axis label
mtext(2, text = expression( ^14*C*-glyphosate line1), line = 3)
mtext(2, text = line2, line = 2)
See ?mtext for more information.
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.
--
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] Questions about sweave... question answered and problem solved
* Dieter Menne wrote, On 2006-08-04 02:57:
Brian Lunergan ff809 at ncf.ca writes:
Evening all:
I'm taking a little time to experiment with R, Sweave, and Miktex/LaTex but
I've run up against some problems and -well- I hope that there are some on
the list who might have some suggestions. This will be kind of wordy as I
will include the complete files involved as I'm just not sure what I'm
looking for. Apologies at the outset.
This is Windows special
See A.12:
http://www.ci.tuwien.ac.at/~leisch/Sweave/FAQ.html
And put a commented (!sic) \usepackage into the header (don't remember where I
found this)
% \usepackage{Sweave}
Dieter
Thanks for the lead. Reinstalled R outside of 'c:\program files'. Reran the
source through Sweave to generate a new example.tex with the preamble:
\documentclass[letterpaper]{article}
\title{Sweave Example 1}
\author{Friedrich Leisch}
\usepackage{C:/R-2.3.1/share/texmf/Sweave}
\begin{document}
Ran it through Latex and everything is as it should be. Problem solved.
--
Brian Lunergan
Nepean, Ontario
Canada
---
avast! Antivirus: Outbound message clean.
Virus Database (VPS): 0631-3, 2006-08-04
Tested on: 2006-08-04 14:52:19
avast! is copyright (c) 2000-2006 ALWIL Software.
http://www.avast.com
__
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] expression() - Superscript in y-axis, keeping line break in string
On Fri, 2006-08-04 at 19:44 +0100, Prof Brian Ripley wrote:
On Fri, 4 Aug 2006, Marc Schwartz (via MN) wrote:
On Fri, 2006-08-04 at 09:47 -0600, Andrew Kniss wrote:
I've tried several different ways to accomplish this, but as yet to no
avail. My y-axis for a plot has a rather long label, and thus I have
been using /n to break it into two lines. However, to make it
technically correct for publication, I also need to use superscript in
the label. For example:
par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1)
plot(1:10,
ylab=14C-glyphosate line1\n line2)
will provide the text in two lines as I would like it. However, I am
trying to keep those same line breaks when using expression() to get my
superscript number. This will not work, as it aligns the 14C section
with the bottom line of the expression making little sense to the
reader.
par(oma=c(0,0,2,0),mar=c(5,6,0.25,2),lheight=1)
plot(1:10,
ylab=expression( ^14*C*-glyphosate line1\n line2))
Is there a way to align the 14C portion of the expression with the top
line of the string rather than the bottom line? Any suggestions are
greatly appreciated.
Andrew
plotmath, as has been covered many times previously, does not support
multi-line expressions. A note should probably be added to ?plotmath on
this.
I've added a note. I think what is exact is that control chars are not
interpreted ('expresssion' is an overloaded work in this context).
Thanks for the nudge (and please do continue to make such remarks).
Brian
snip
Happy to help and thanks for both noticing and taking the time to
incorporate the update.
Best regards,
Marc
__
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] training svm's with probability flag (re-send in plain text)
Hi- I'm seeing some weirdness with svm and tune.svm that I can't figure out- was wondering if anyone else has seen this? Perhaps I'm failing to make something the expected class? Below is my repro case, though it *sometimes* doesn't repro. I'm using R2.3.1 on WindowsXP. I was also seeing it happen with R2.1.1 and have seen it on 2 different machines. data(iris) attach(iris) library(e1071) train- iris[c(1:30,50:80,100:130),] test- iris[-c(1:30,50:80,100:130),] y.train- train$Species y.test- test$Species obj- tune.svm(train[,-5], y.train, gamma = 2^(-1:1), cost = 2^(2:4), probability=T) my.svm- obj$best.model pred1- predict(my.svm, test[,-5]) pred2- predict(my.svm, test[,-5], probability=T) table(pred1, y.test) table(pred2, y.test) When I do this, the two different tables often come out different, as below: table(pred1, y.test) y.test pred1setosa versicolor virginica setosa 19 0 0 versicolor 0 18 1 virginica 0 119 table(pred2, y.test) y.test pred2setosa versicolor virginica setosa 18 0 0 versicolor 1 18 1 virginica 0 119 I'm not sure 1. why the results would differ based on whether I choose to calculate the probabilities, and 2. which one to trust?? Anyone come across this before, or have any ideas? thanks, jessie __ 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] Doubt about Student t distribution simulation
Dear R list,
I would like to illustrate the origin of the Student t distribution using R.
So, if (sample.mean - pop.mean) / standard.error(sample.mean) has t
distribution with (sample.size - 1) degree free, what is wrong with the
simulation below? I think that the theoretical curve should agree with
the relative frequencies of the t values calculated:
#== begin options=
# parameters
mu= 10
sigma = 5
# size of sample
n = 3
# repetitions
nsim = 1
# histogram parameter
nchist = 150
#== end options===
t = numeric()
pop = rnorm(1, mean = mu, sd = sigma)
for (i in 1:nsim) {
amo.i = sample(pop, n, replace = TRUE)
t[i] = (mean(amo.i) - mu) / (sigma / sqrt(n))
}
win.graph(w = 5, h = 7)
split.screen(c(2,1))
screen(1)
hist(t,
main = histogram,
breaks = nchist,
col = lightgray,
xlab = '', ylab = Fi,
font.lab = 2, font = 2)
screen(2)
hist(t,
probability = T,
main= 'f.d.p and histogram',
breaks = nchist,
col = 'lightgray',
xlab= 't', ylab = 'f(t)',
font.lab= 2, font = 2)
x = t
curve(dt(x, df = n-1), add = T, col = red, lwd = 2)
Many thanks for any help,
___
Jose Claudio Faria
Brasil/Bahia/Ilheus/UESC/DCET
Estatística Experimental/Prof. Adjunto
mails: [EMAIL PROTECTED]
[EMAIL PROTECTED]
[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] meta characters in file path
thanks Prof Ripley. dir() returns the path with full names (wildcards replaced) that are exactly what I need. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Prof Brian Ripley Sent: Friday, August 04, 2006 12:27 PM To: Li,Qinghong,ST.LOUIS,Molecular Biology Cc: r-help@stat.math.ethz.ch; Tony Plate Subject: Re: [R] meta characters in file path On Fri, 4 Aug 2006, Li,Qinghong,ST.LOUIS,Molecular Biology wrote: Thanks. I tried them, it works for most of those characters except * and ?. Those are not valid characters in Windows file paths (/ * : ? \ | are invalid in file or dir names). Does regular expression work in file names in windows? No, and I think you may mean wildcards (which is what work on the command line). e.g. I have a machine-generated file named 021706 matrix#1479 @50.csv, of which 1479 is kinda random. Will I be able to match 1479 with some sort of wild card chars? Yes, use dir(), with regexp pattern patching to find the name(s) you want. glob2rx() might be useful here. Thanks Johnny -Original Message- From: Tony Plate [mailto:[EMAIL PROTECTED] Sent: Thursday, August 03, 2006 3:42 PM To: Li,Qinghong,ST.LOUIS,Molecular Biology Cc: r-help@stat.math.ethz.ch Subject: Re: [R] meta characters in file path What is the problem you are having? Seems to work fine for me running under Windows2000: write.table(data.frame(a=1:3,b=4:6), file=@# x.csv, sep=,) read.csv(file=@# x.csv) a b 1 1 4 2 2 5 3 3 6 sessionInfo() Version 2.3.1 (2006-06-01) i386-pc-mingw32 attached base packages: [1] methods stats graphics grDevices utils datasets [7] base other attached packages: XML 0.99-8 Li,Qinghong,ST.LOUIS,Molecular Biology wrote: Hi, I need to read in some files. The file names contain come meta characters such as @, #, and white spaces etc, In read.csv, file= option, is there any way that one can make the function to recognize a file path with those characters? Thanks Johnny [[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. -- 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.
Re: [R] Doubt about Student t distribution simulation
Dear Jose,
The problem is that you're using the population standard deviation (sigma)
rather than the sample SD of each sample [i.e., t[i] = (mean(amo.i) - mu) /
(sd(amo.i) / sqrt(n)) ], so your values should be normally distributed, as
they appear to be.
A couple of smaller points: (1) Even after this correction, you're sampling
from a discrete population (albeit with replacement) and so the values won't
be exactly t-distributed. You could draw the samples directly from N(mu,
sigma) instead. (2) It would be preferable to make a quantile-comparison
plot against the t-distribution, since you'd get a better picture of what's
going on in the tails.
I hope this helps,
John
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario
Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Jose
Claudio Faria
Sent: Friday, August 04, 2006 3:09 PM
To: R-help@stat.math.ethz.ch
Subject: [R] Doubt about Student t distribution simulation
Dear R list,
I would like to illustrate the origin of the Student t
distribution using R.
So, if (sample.mean - pop.mean) / standard.error(sample.mean)
has t distribution with (sample.size - 1) degree free, what
is wrong with the simulation below? I think that the
theoretical curve should agree with the relative frequencies
of the t values calculated:
#== begin options=
# parameters
mu= 10
sigma = 5
# size of sample
n = 3
# repetitions
nsim = 1
# histogram parameter
nchist = 150
#== end options===
t = numeric()
pop = rnorm(1, mean = mu, sd = sigma)
for (i in 1:nsim) {
amo.i = sample(pop, n, replace = TRUE)
t[i] = (mean(amo.i) - mu) / (sigma / sqrt(n)) }
win.graph(w = 5, h = 7)
split.screen(c(2,1))
screen(1)
hist(t,
main = histogram,
breaks = nchist,
col = lightgray,
xlab = '', ylab = Fi,
font.lab = 2, font = 2)
screen(2)
hist(t,
probability = T,
main= 'f.d.p and histogram',
breaks = nchist,
col = 'lightgray',
xlab= 't', ylab = 'f(t)',
font.lab= 2, font = 2)
x = t
curve(dt(x, df = n-1), add = T, col = red, lwd = 2)
Many thanks for any help,
___
Jose Claudio Faria
Brasil/Bahia/Ilheus/UESC/DCET
Estatística Experimental/Prof. Adjunto
mails: [EMAIL PROTECTED]
[EMAIL PROTECTED]
[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.
__
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] Doubt about Student t distribution simulation
Jose Claudio Faria [EMAIL PROTECTED] writes:
Dear R list,
I would like to illustrate the origin of the Student t distribution using R.
So, if (sample.mean - pop.mean) / standard.error(sample.mean) has t
distribution with (sample.size - 1) degree free, what is wrong with the
simulation below? I think that the theoretical curve should agree with
the relative frequencies of the t values calculated:
#== begin options=
# parameters
mu= 10
sigma = 5
# size of sample
n = 3
# repetitions
nsim = 1
# histogram parameter
nchist = 150
#== end options===
t = numeric()
pop = rnorm(1, mean = mu, sd = sigma)
for (i in 1:nsim) {
amo.i = sample(pop, n, replace = TRUE)
t[i] = (mean(amo.i) - mu) / (sigma / sqrt(n))
At the very least, you need a sample-based standard error: sd(amo.i),
not sigma. Also, resampling from pop is not really what the
t-distribution is based on, but I don't think that matters much.
}
win.graph(w = 5, h = 7)
split.screen(c(2,1))
screen(1)
hist(t,
main = histogram,
breaks = nchist,
col = lightgray,
xlab = '', ylab = Fi,
font.lab = 2, font = 2)
screen(2)
hist(t,
probability = T,
main= 'f.d.p and histogram',
breaks = nchist,
col = 'lightgray',
xlab= 't', ylab = 'f(t)',
font.lab= 2, font = 2)
x = t
curve(dt(x, df = n-1), add = T, col = red, lwd = 2)
Many thanks for any help,
___
Jose Claudio Faria
Brasil/Bahia/Ilheus/UESC/DCET
Estatística Experimental/Prof. Adjunto
mails: [EMAIL PROTECTED]
[EMAIL PROTECTED]
[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.
--
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.
Re: [R] Error when loading odesolve
Prof. Ripley: Thanks for diagnosing Wuming Gong's problem. I'm not sure I would have recognized the solution so quickly. I have uploaded to CRAN a new version of odesolve with Makevars fixed. R. Woodrow Setzer, Ph. D. National Center for Computational Toxicology US Environmental Protection Agency Mail Drop B205-01/US EPA/RTP, NC 27711 Ph: (919) 541-0128Fax: (919) 541-1194 Prof Brian Ripley [EMAIL PROTECTED]To .ac.uk Wuming Gong Sent by: [EMAIL PROTECTED] [EMAIL PROTECTED]cc tat.math.ethz.ch r-help@stat.math.ethz.ch Subject Re: [R] Error when loading 08/04/2006 01:33 odesolve PM Looking at odesolve, src/Makevars is PKG_LIBS=$(BLAS_LIBS) Now, the documentation says that if you have $(BLAS_LIBS) you must also have $(FLIBS), so please change this to PKG_LIBS=$(BLAS_LIBS) $(FLIBS) and take this up with the package maintainer (which is what the posting guide asked you to do in the first place). On Sat, 5 Aug 2006, Wuming Gong wrote: Dear list, I installed odesolve package (0.5-15) in R 2.3.1 in a Solaris server (Generic_118558-11 sun4u sparc SUNW,Sun-Blade-1000). The installing progress completed without errors, though several warnings like Warning: Option -fPIC passed to ld, if ld is invoked, ignored otherwise were outputed. However, when loading the odesolve package by library(odesolve), following error messages pop out: library(odesolve) Error in dyn.load(x, as.logical(local), as.logical(now)) : unable to load shared library '/project/scratch/ligroup/R1/lib/R/library/odesolve/libs/odesolve.so': ld.so.1: R: fatal: relocation error: file /project/scratch/ligroup/R1/lib/R/library/odesolve/libs/odesolve.so: symbol __f90_ssfw: referenced symbol not found Error: package/namespace load failed for 'odesolve' Could any one tell me how to fix this problem? Thanks very much. Wuming __ 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-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] Doubt about Student t distribution simulation
Hi, Jose/John,
Here's an example to help Jose and highlights John's advice. Also
includes set.seed which should be included in all simulations posted to
R-help.
set.seed(42)
mu - 10
sigma - 5
n - 3
nsim - 1
m - matrix(rnorm(n * nsim, mu, sigma), nsim, n)
t - apply(m, 1, function(x) (mean(x) - mu)/(sd(x)/sqrt(n)))
library(lattice)
qqmath(t, distribution = function(x) qt(x, n - 1),
panel = function(x, ...) {
panel.qqmath(x, col = darkblue, ...)
panel.qqmathline(x, col = darkred, ...)
})
With n = 3, expect a few outliers.
--sundar
John Fox wrote:
Dear Jose,
The problem is that you're using the population standard deviation (sigma)
rather than the sample SD of each sample [i.e., t[i] = (mean(amo.i) - mu) /
(sd(amo.i) / sqrt(n)) ], so your values should be normally distributed, as
they appear to be.
A couple of smaller points: (1) Even after this correction, you're sampling
from a discrete population (albeit with replacement) and so the values won't
be exactly t-distributed. You could draw the samples directly from N(mu,
sigma) instead. (2) It would be preferable to make a quantile-comparison
plot against the t-distribution, since you'd get a better picture of what's
going on in the tails.
I hope this helps,
John
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario
Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Jose
Claudio Faria
Sent: Friday, August 04, 2006 3:09 PM
To: R-help@stat.math.ethz.ch
Subject: [R] Doubt about Student t distribution simulation
Dear R list,
I would like to illustrate the origin of the Student t
distribution using R.
So, if (sample.mean - pop.mean) / standard.error(sample.mean)
has t distribution with (sample.size - 1) degree free, what
is wrong with the simulation below? I think that the
theoretical curve should agree with the relative frequencies
of the t values calculated:
#== begin options=
# parameters
mu= 10
sigma = 5
# size of sample
n = 3
# repetitions
nsim = 1
# histogram parameter
nchist = 150
#== end options===
t = numeric()
pop = rnorm(1, mean = mu, sd = sigma)
for (i in 1:nsim) {
amo.i = sample(pop, n, replace = TRUE)
t[i] = (mean(amo.i) - mu) / (sigma / sqrt(n)) }
win.graph(w = 5, h = 7)
split.screen(c(2,1))
screen(1)
hist(t,
main = histogram,
breaks = nchist,
col = lightgray,
xlab = '', ylab = Fi,
font.lab = 2, font = 2)
screen(2)
hist(t,
probability = T,
main= 'f.d.p and histogram',
breaks = nchist,
col = 'lightgray',
xlab= 't', ylab = 'f(t)',
font.lab= 2, font = 2)
x = t
curve(dt(x, df = n-1), add = T, col = red, lwd = 2)
Many thanks for any help,
___
Jose Claudio Faria
Brasil/Bahia/Ilheus/UESC/DCET
Estatística Experimental/Prof. Adjunto
mails: [EMAIL PROTECTED]
[EMAIL PROTECTED]
[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.
__
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] Doubt about Student t distribution simulation
Dears John, Peter and Sundar,
Many thanks for the quick answers!!!
.. and sorry for all..
[]s
___
Jose Claudio Faria
Brasil/Bahia/Ilheus/UESC/DCET
Estatística Experimental/Prof. Adjunto
mails: [EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
John Fox escreveu:
Dear Jose,
The problem is that you're using the population standard deviation (sigma)
rather than the sample SD of each sample [i.e., t[i] = (mean(amo.i) - mu) /
(sd(amo.i) / sqrt(n)) ], so your values should be normally distributed, as
they appear to be.
A couple of smaller points: (1) Even after this correction, you're sampling
from a discrete population (albeit with replacement) and so the values won't
be exactly t-distributed. You could draw the samples directly from N(mu,
sigma) instead. (2) It would be preferable to make a quantile-comparison
plot against the t-distribution, since you'd get a better picture of what's
going on in the tails.
I hope this helps,
John
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario
Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Jose
Claudio Faria
Sent: Friday, August 04, 2006 3:09 PM
To: R-help@stat.math.ethz.ch
Subject: [R] Doubt about Student t distribution simulation
Dear R list,
I would like to illustrate the origin of the Student t
distribution using R.
So, if (sample.mean - pop.mean) / standard.error(sample.mean)
has t distribution with (sample.size - 1) degree free, what
is wrong with the simulation below? I think that the
theoretical curve should agree with the relative frequencies
of the t values calculated:
#== begin options=
# parameters
mu= 10
sigma = 5
# size of sample
n = 3
# repetitions
nsim = 1
# histogram parameter
nchist = 150
#== end options===
t = numeric()
pop = rnorm(1, mean = mu, sd = sigma)
for (i in 1:nsim) {
amo.i = sample(pop, n, replace = TRUE)
t[i] = (mean(amo.i) - mu) / (sigma / sqrt(n)) }
win.graph(w = 5, h = 7)
split.screen(c(2,1))
screen(1)
hist(t,
main = histogram,
breaks = nchist,
col = lightgray,
xlab = '', ylab = Fi,
font.lab = 2, font = 2)
screen(2)
hist(t,
probability = T,
main= 'f.d.p and histogram',
breaks = nchist,
col = 'lightgray',
xlab= 't', ylab = 'f(t)',
font.lab= 2, font = 2)
x = t
curve(dt(x, df = n-1), add = T, col = red, lwd = 2)
Many thanks for any help,
___
Jose Claudio Faria
Brasil/Bahia/Ilheus/UESC/DCET
Estatística Experimental/Prof. Adjunto
mails: [EMAIL PROTECTED]
[EMAIL PROTECTED]
[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.
Esta mensagem foi verificada pelo E-mail Protegido Terra.
Scan engine: McAfee VirusScan / Atualizado em 04/08/2006 / Versão: 4.4.00/4822
Proteja o seu e-mail Terra: http://mail.terra.com.br/
__
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] Doubt about Student t distribution simulation
Dear Sundar,
Try qq.plot(t, dist=t, df=n-1) from the car package, which include a
95-percent point-wise confidence envelope that helps you judge how extreme
the outliers are relative to expectations.
Regards,
John
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario
Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox
-Original Message-
From: Sundar Dorai-Raj [mailto:[EMAIL PROTECTED]
Sent: Friday, August 04, 2006 3:27 PM
To: John Fox
Cc: [EMAIL PROTECTED]; R-help@stat.math.ethz.ch
Subject: Re: [R] Doubt about Student t distribution simulation
Hi, Jose/John,
Here's an example to help Jose and highlights John's advice.
Also includes set.seed which should be included in all
simulations posted to R-help.
set.seed(42)
mu - 10
sigma - 5
n - 3
nsim - 1
m - matrix(rnorm(n * nsim, mu, sigma), nsim, n) t -
apply(m, 1, function(x) (mean(x) - mu)/(sd(x)/sqrt(n)))
library(lattice)
qqmath(t, distribution = function(x) qt(x, n - 1),
panel = function(x, ...) {
panel.qqmath(x, col = darkblue, ...)
panel.qqmathline(x, col = darkred, ...)
})
With n = 3, expect a few outliers.
--sundar
John Fox wrote:
Dear Jose,
The problem is that you're using the population standard deviation
(sigma) rather than the sample SD of each sample [i.e., t[i] =
(mean(amo.i) - mu) /
(sd(amo.i) / sqrt(n)) ], so your values should be normally
distributed, as they appear to be.
A couple of smaller points: (1) Even after this correction, you're
sampling from a discrete population (albeit with
replacement) and so
the values won't be exactly t-distributed. You could draw
the samples
directly from N(mu,
sigma) instead. (2) It would be preferable to make a
quantile-comparison plot against the t-distribution, since
you'd get a
better picture of what's going on in the tails.
I hope this helps,
John
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario
Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Jose Claudio
Faria
Sent: Friday, August 04, 2006 3:09 PM
To: R-help@stat.math.ethz.ch
Subject: [R] Doubt about Student t distribution simulation
Dear R list,
I would like to illustrate the origin of the Student t distribution
using R.
So, if (sample.mean - pop.mean) / standard.error(sample.mean) has t
distribution with (sample.size - 1) degree free, what is wrong with
the simulation below? I think that the theoretical curve
should agree
with the relative frequencies of the t values calculated:
#== begin options=
# parameters
mu= 10
sigma = 5
# size of sample
n = 3
# repetitions
nsim = 1
# histogram parameter
nchist = 150
#== end options===
t = numeric()
pop = rnorm(1, mean = mu, sd = sigma)
for (i in 1:nsim) {
amo.i = sample(pop, n, replace = TRUE)
t[i] = (mean(amo.i) - mu) / (sigma / sqrt(n)) }
win.graph(w = 5, h = 7)
split.screen(c(2,1))
screen(1)
hist(t,
main = histogram,
breaks = nchist,
col = lightgray,
xlab = '', ylab = Fi,
font.lab = 2, font = 2)
screen(2)
hist(t,
probability = T,
main= 'f.d.p and histogram',
breaks = nchist,
col = 'lightgray',
xlab= 't', ylab = 'f(t)',
font.lab= 2, font = 2)
x = t
curve(dt(x, df = n-1), add = T, col = red, lwd = 2)
Many thanks for any help,
___
Jose Claudio Faria
Brasil/Bahia/Ilheus/UESC/DCET
Estatística Experimental/Prof. Adjunto
mails: [EMAIL PROTECTED]
[EMAIL PROTECTED]
[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.
__
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] Help with short time series
Dear R-list, I have a statistical problem with the comparison of two short time-series of density data in an ecological framework. I have to compare two short time series (5 years, one value for each year) of species density data (it is the density of fish in two different streams) to test if the two means of the five densities are significantly different, so basically if the two mean stream-specific fish densities are significantly different. I don't think I can use a straight t-test due to the problem of autocorrelation and I don't think I can use a repeated measure ANOVA as I don't have any replicates. Any help would be greatly appreciated. Thanks Simone Vincenzi -- Universita' degli Studi di Parma (http://www.unipr.it) __ 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] Postscript fonts
How can I find out what fonts are available for par(family= for the postscript device? -- Erich Neuwirth, Didactic Center for Computer Science University of Vienna Visit our SunSITE at http://sunsite.univie.ac.at Phone: +43-1-4277-39464 Fax: +43-1-4277-9394 __ 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] Variance-Covariance matrix from glm()
We are trying to find out how to get the variance-covariance matrix of the MLEs out of the glm function. Can anyone help? Thanks, Daniel Jeske Department of Statistics University of California Riverside, CA [[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] polychoric correlation error
Dear all,
I get a strange error when I find polychoric correlations with the ML method,
which I have been able to reproduce using randomly-generated data.
What is wrong?
I realize that the data that I generated randomly is a bit strange, but it is
the only way that I duplicate the error message.
n-100
test.x-rnorm(n, mean=0, sd=1)
test.c-test.x + rnorm(n, mean=0, sd=.5)
thresh.x-c(-2.5, -1, -.5, .5, 1000)
thresh.c-c(-1, 1, 2, 3, 1000)
discrete.x-discrete.c-vector(length=n)
for (i in 1:n) {
+ discrete.x[i]-which.min(thresh.x test.x[i] )
+ discrete.c[i]-which.min(thresh.c test.c[i] )
+ }
pc-polychor(discrete.x, discrete.c, std.err=T, ML=T)
Error in optim(c(optimise(f, interval = c(-1, 1))$minimum, rc, cc), f, :
non-finite finite-difference value [1]
In addition: There were 50 or more warnings (use warnings() to see the first 50)
print(pc)
Error in print(pc) : object pc not found
warnings()
Warning messages:
1: NaNs produced in: log(x)
2: NA/Inf replaced by maximum positive value
3: NaNs produced in: log(x)
---
Thanks,
Janet
This email message is for the sole use of the intended recip...{{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] Variance-Covariance matrix from glm()
On Fri, 4 Aug 2006, Daniel Jeske wrote: We are trying to find out how to get the variance-covariance matrix of the MLEs out of the glm function. Can anyone help? It can be extracted with the corresponding vcov() method. Best, Z __ 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] (... not defined because of singularities) in lm()
I got the message, Coefficients: (1 not defined because of singularities), in the returned result of lm(). What does it mean? And where should I start investigating why it happens? thanks. /// Complete result of lm() Call: lm(formula = durationRatio ~ isHStar + isWordFinal + oneSyllToEOW + isInterNuclear + isIntonNuclear + isInterLP + isIntonLF + zeroSyllToNA + oneSyllToNA + zeroSyllToEOP + oneSyllToEOP + isInterNuclear:zeroSyllToEOP) Residuals: Min1QMedian3Q Max -3.031538 -0.126943 -0.002909 0.121606 4.322135 Coefficients: (1 not defined because of singularities) Estimate Std. Error t value Pr(|t|) (Intercept) 1.044481 0.010694 97.669 2e-16 *** isHStar -0.055227 0.005748 -9.607 2e-16 *** isWordFinal -0.262071 0.009982 -26.254 2e-16 *** oneSyllToEOW -0.031313 0.009917 -3.158 0.00160 ** isInterNuclear0.063470 0.015699 4.043 5.32e-05 *** isIntonNuclear0.054570 0.011711 4.660 3.21e-06 *** isInterLP-0.262302 0.011868 -22.102 2e-16 *** isIntonLF-0.160018 0.012537 -12.764 2e-16 *** zeroSyllToNA -0.160767 0.011105 -14.477 2e-16 *** oneSyllToNA -0.012633 0.010779 -1.172 0.24123 zeroSyllToEOP-0.212474 0.011987 -17.726 2e-16 *** oneSyllToEOP -0.123324 0.010998 -11.214 2e-16 *** isInterNuclear:zeroSyllToEOPNA NA NA NA --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 0.2662 on 9650 degrees of freedom Multiple R-Squared: 0.3957, Adjusted R-squared: 0.395 F-statistic: 574.5 on 11 and 9650 DF, p-value: 2.2e-16 __ 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] cor of two matrices whose columns got shuffled
Hello, I'm trying to take the correlation of two distance matrices and get the r2 value ex) Ndisc PS1Ntet Nsito NcB NcA PS3 NcC NiB NiA PS1 0.08945 Ntet 0.08601 0.02020 Nsito 0.09361 0.02780 0.02260 NcB 0.09479 0.03258 0.02914 0.03674 NcA 0.09363 0.03142 0.02798 0.03558 0.02190 PS3 0.09145 0.02924 0.02580 0.03340 0.02446 0.02330 NcC 0.08919 0.02698 0.02354 0.03114 0.02220 0.02104 0.01670 NiB 0.09749 0.03802 0.03458 0.04218 0.04336 0.04220 0.04002 0.03776 NiA 0.09571 0.03624 0.03280 0.04040 0.04158 0.04042 0.03824 0.03598 0.02170 PS2 0.09234 0.03287 0.02943 0.03703 0.03821 0.03705 0.03487 0.03261 0.03289 0.03111 and Ndisc PS1 Ntet NcB NcA NcC PS3 Nsito PS2 NiB PS1 2.981494 Ntet 2.949046 1.003718 NcB 3.212256 1.266928 1.178326 NcA 3.531725 1.586397 1.497795 1.217055 NcC 3.111421 1.166093 1.077491 0.944637 1.264106 PS3 3.115273 1.169945 1.081343 1.000991 1.320460 0.900156 Nsito 2.994152 1.131330 1.098882 1.362092 1.681561 1.261257 1.265109 PS2 2.843805 1.228361 1.195913 1.459123 1.778592 1.358288 1.362140 1.241019 NiB 2.805537 1.375615 1.343167 1.606377 1.925846 1.505542 1.509394 1.388273 1.237926 NiA 2.752269 1.322347 1.289899 1.553109 1.872578 1.452274 1.456126 1.335005 1.184658 1.018210 The problem is that these matrices are the endpoint of an extensive simulation involving phylogenetic trees and the order of the columns and rows shifts around according to how they were ordered in the tree. I used cophenetic.phylo to get the distance matrix and now I need to find a way to reorder or sort the columns and rows such that the two matrices match and cor is taking the correlation of the corresponding column and row and not just the one in the same position. I don't want to sort the values in the columns and rows but the entire column and their associated rows. I've looked at order , sort, and match and can't figure out a way to do it based on the alphanumeric name of the columns. Does anyone have any suggestions? Sincerely, Betty -- Betty Gilbert [EMAIL PROTECTED] Taylor Lab Plant and Microbial Biology 321 Koshland Hall U.C. Berkeley Berkeley, Ca 94720 [[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] polychoric correlation error
Dear Janet,
Because you didn't set the value of the random-number generator seed, your
example isn't precisely reproducible, but the problem is apparent anyway:
set.seed(12345)
n-100
test.x-rnorm(n, mean=0, sd=1)
test.c-test.x + rnorm(n, mean=0, sd=.5)
thresh.x-c(-2.5, -1, -.5, .5, 1000)
thresh.c-c(-1, 1, 2, 3, 1000)
discrete.x-discrete.c-vector(length=n)
for (i in 1:n) {
+ discrete.x[i]-which.min(thresh.x test.x[i] )
+ discrete.c[i]-which.min(thresh.c test.c[i] ) }
table(discrete.x, discrete.c)
discrete.c
discrete.x 1 2 3 4 5
2 12 1 0 0 0
3 3 12 0 0 0
4 2 19 2 0 0
5 0 18 21 9 1
cor(test.x, test.c)
[1] 0.9184189
pc - polychor(discrete.x, discrete.c, std.err=T, ML=T)
Warning messages:
1: NaNs produced in: log(x)
2: NaNs produced in: log(x)
3: NaNs produced in: log(x)
pc
Polychoric Correlation, ML est. = 0.9077 (0.03314)
Test of bivariate normality: Chisquare = 3.103, df = 11, p = 0.9893
Row Thresholds
Threshold Std.Err.
1 -1.12200 0.1609
2 -0.56350 0.1309
3 0.03318 0.1235
Column Thresholds
Threshold Std.Err.
1 -0.9389 0.1489
20.4397 0.1292
31.2790 0.1707
42.3200 0.3715
The variables that you've created are indeed bivariate normal, but they are
highly correlated, and your choice of cut points makes it hard to estimate
the correlation from the contingency tables, apparently producing some
difficulty in the maximization of the likelihood. Nevertheless, the ML
estimates of the correlation and thresholds for the set of data above are
pretty good. (In your case, the optimization failed.)
BTW, a more straightforward way to create the categorical variables would be
discrete.x - cut(test.x, c(-Inf, -2.5, -1, -.5, .5, Inf))
discrete.c - cut(test.c, c(-Inf, -1, 1, 2, 3, Inf))
I hope this helps,
John
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario
Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Rosenbaum, Janet
Sent: Friday, August 04, 2006 5:49 PM
To: r-help@stat.math.ethz.ch
Subject: [R] polychoric correlation error
Dear all,
I get a strange error when I find polychoric correlations
with the ML method, which I have been able to reproduce using
randomly-generated data.
What is wrong?
I realize that the data that I generated randomly is a bit
strange, but it is the only way that I duplicate the error message.
n-100
test.x-rnorm(n, mean=0, sd=1)
test.c-test.x + rnorm(n, mean=0, sd=.5) thresh.x-c(-2.5, -1, -.5,
.5, 1000) thresh.c-c(-1, 1, 2, 3, 1000)
discrete.x-discrete.c-vector(length=n)
for (i in 1:n) {
+ discrete.x[i]-which.min(thresh.x test.x[i] )
+ discrete.c[i]-which.min(thresh.c test.c[i] ) }
pc-polychor(discrete.x, discrete.c, std.err=T, ML=T)
Error in optim(c(optimise(f, interval = c(-1, 1))$minimum,
rc, cc), f, :
non-finite finite-difference value [1]
In addition: There were 50 or more warnings (use warnings()
to see the first 50)
print(pc)
Error in print(pc) : object pc not found
warnings()
Warning messages:
1: NaNs produced in: log(x)
2: NA/Inf replaced by maximum positive value
3: NaNs produced in: log(x)
---
Thanks,
Janet
This email message is for the sole use of the intended\ ...{{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.
