Re: [R] (no subject)
type ?par and then have a look at: cex.lab, cex.main cheers christoph [EMAIL PROTECTED] wrote: Dear ladies and gentlemen! When I use the plot funtion how can I change the size of the title for the x and y axes (xlab, ylab)and the size of the axes label ? Thank you very much. With best regards Claudia __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Strange behaviour of load
On Wed, 11 Jan 2006, giovanni parrinello wrote:
Dear All,
simetimes when I load an Rdata I get this message
###
Code:
load('bladder1.RData')
Carico il pacchetto richiesto: rpart ( Bad traslastion: Load required
package-...)
Carico il pacchetto richiesto: MASS
Carico il pacchetto richiesto: mlbench
Carico il pacchetto richiesto: survival
Carico il pacchetto richiesto: splines
Carico il pacchetto richiesto: 'survival'
The following object(s) are masked from package:Hmisc :
untangle.specials
Carico il pacchetto richiesto: class
Carico il pacchetto richiesto: nnet
#
So I have many unrequired packages loaded.
Any idea?
They are required! My guess is that you have object(s) saved with
environment the namespace of some package, and loading that namespace is
pulling these in. The only CRAN package which requires mlbench appears to
be ipred, and that requires all of those except splines, required by
survival.
So I believe you have been using ipred and have saved a reference to its
namespace.
--
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
[R] dataframes with only one variable
Subsetting from a dataframe with only one variable returns a vector, not a dataframe. This seems somewhat inconsistent. Wouldn't it be better if subsetting would respect the structure completely? v1-1:4 v2-4:1 df1-data.frame(v1) df2-data.frame(v1,v2) sel1-c(TRUE,TRUE,TRUE,TRUE) df1[sel1,] [1] 1 2 3 4 df2[sel1,] v1 v2 1 1 4 2 2 3 3 3 2 4 4 1 -- Erich Neuwirth Institute for Scientific Computing and Didactic Center for Computer Science University of Vienna phone: +43-1-4277-39464 fax: +43-1-4277-39459 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] dataframes with only one variable
Subsetting from a dataframe with only one variable returns a vector, not a dataframe. This seems somewhat inconsistent. Wouldn't it be better if subsetting would respect the structure completely? v1-1:4 v2-4:1 df1-data.frame(v1) df2-data.frame(v1,v2) sel1-c(TRUE,TRUE,TRUE,TRUE) df1[sel1,] df1[[sel1, , drop=FALSE] Should do what you want. Best, Matthias [1] 1 2 3 4 df2[sel1,] v1 v2 1 1 4 2 2 3 3 3 2 4 4 1 -- Erich Neuwirth Institute for Scientific Computing and Didactic Center for Computer Science University of Vienna phone: +43-1-4277-39464 fax: +43-1-4277-39459 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] F-test degree of freedoms in lme4 ?
Dear Wilhelm There is an article, including a part about fitting linear mixed models. There the problem with the degrees of freedom is described. You can have a look to the second link, too, discussing the problem as well. http://cran.r-project.org/doc/Rnews/Rnews_2005-1.pdf http://finzi.psych.upenn.edu/R/Rhelp02a/archive/67414.html Regards, Christoph Buser -- Christoph Buser [EMAIL PROTECTED] Seminar fuer Statistik, LEO C13 ETH (Federal Inst. Technology) 8092 Zurich SWITZERLAND phone: x-41-44-632-4673 fax: 632-1228 http://stat.ethz.ch/~buser/ -- Wilhelm B. Kloke writes: I have a problem moving from multistratum aov analysis to lmer. My dataset has observations of ampl at 4 levels of gapf and 2 levels of bl on 6 subjects levels VP, with 2 replicates wg each, and is balanced. Here is the summary of this set with aov: summary(aov(ampl~gapf*bl+Error(VP/(bl*gapf)),hframe2)) Error: VP Df Sum Sq Mean Sq F value Pr(F) Residuals 5531 106 Error: VP:bl Df Sum Sq Mean Sq F value Pr(F) bl 1 1700170037.8 0.0017 ** Residuals 5225 45 --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 Error: VP:gapf Df Sum Sq Mean Sq F value Pr(F) gapf 3933 31124.2 5.3e-06 *** Residuals 15193 13 --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 Error: VP:bl:gapf Df Sum Sq Mean Sq F value Pr(F) gapf:bl3 93.931.33.68 0.036 * Residuals 15 127.6 8.5 --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 Error: Within Df Sum Sq Mean Sq F value Pr(F) Residuals 48318 7 This is mostly identical the analysis by BMDP 4V, except for the Greenhouse-Geisser epsilons, which are not estimated this way. I have to analyse a similar dataset, which is not balanced. So I need to change the method. Following Pinheiro/Bates p.90f, I tried hf2.lme - lme(ampl~gapf*bl,hframe2,random=list(VP=pdDiag(~gapf*bl),bl=pdDiag(~gapf))) and some variations of this to get the same F tests generated. At least, I got the F-test on error stratum VP:bl this way, but not the other two: anova(hf2.lme) numDF denDF F-value p-value (Intercept) 178 764.86 .0001 gapf378 17.68 .0001 bl 1 5 37.81 0.0017 gapf:bl 3782.99 0.0362 Then I tried to move to lmer. I tried to find something equivalent to the above lme call, with no success at all. In case, that the problem is in the data, here is the set: VP ampl wg bl gapf 1 WJ 22 w s 144 2 CR 23 w s 144 3 MZ 25 w s 144 4 MP 34 w s 144 5 HJ 36 w s 144 6 SJ 26 w s 144 7 WJ 34 w s 80 8 CR 31 w s 80 9 MZ 33 w s 80 10 MP 36 w s 80 11 HJ 37 w s 80 12 SJ 32 w s 80 13 WJ 34 w s 48 14 CR 37 w s 48 15 MZ 38 w s 48 16 MP 38 w s 48 17 HJ 40 w s 48 18 SJ 32 w s 48 19 WJ 36 w s 16 20 CR 40 w s 16 21 MZ 39 w s 16 22 MP 40 w s 16 23 HJ 40 w s 16 24 SJ 38 w s 16 25 WJ 16 g s 144 26 CR 28 g s 144 27 MZ 18 g s 144 28 MP 33 g s 144 29 HJ 37 g s 144 30 SJ 28 g s 144 31 WJ 28 g s 80 32 CR 33 g s 80 33 MZ 24 g s 80 34 MP 34 g s 80 35 HJ 36 g s 80 36 SJ 30 g s 80 37 WJ 32 g s 48 38 CR 38 g s 48 39 MZ 34 g s 48 40 MP 37 g s 48 41 HJ 39 g s 48 42 SJ 30 g s 48 43 WJ 36 g s 16 44 CR 34 g s 16 45 MZ 36 g s 16 46 MP 40 g s 16 47 HJ 40 g s 16 48 SJ 36 g s 16 49 WJ 22 w b 144 50 CR 24 w b 144 51 MZ 20 w b 144 52 MP 26 w b 144 53 HJ 22 w b 144 54 SJ 16 w b 144 55 WJ 26 w b 80 56 CR 24 w b 80 57 MZ 26 w b 80 58 MP 27 w b 80 59 HJ 26 w b 80 60 SJ 18 w b 80 61 WJ 28 w b 48 62 CR 23 w b 48 63 MZ 28 w b 48 64 MP 29 w b 48 65 HJ 27 w b 48 66 SJ 24 w b 48 67 WJ 32 w b 16 68 CR 26 w b 16 69 MZ 30 w b 16 70 MP 28 w b 16 71 HJ 30 w b 16 72 SJ 22 w b 16 73 WJ 22 g b 144 74 CR 18 g b 144 75 MZ 18 g b 144 76 MP 26 g b 144 77 HJ 22 g b 144 78 SJ 18 g b 144 79 WJ 24 g b 80 80 CR 26 g b 80 81 MZ 30 g b 80 82 MP 26 g b 80 83 HJ 26 g b 80 84 SJ 24 g b 80 85 WJ 28 g b 48 86 CR 28 g b 48 87 MZ 27 g b 48 88 MP 30 g b 48 89 HJ 26 g b 48 90 SJ 16 g b 48 91 WJ 28 g b 16 92 CR 19 g b 16 93 MZ 24 g b 16 94 MP 32 g b 16 95 HJ 30 g b 16 96 SJ 22 g b 16 -- Dipl.-Math. Wilhelm Bernhard Kloke Institut fuer Arbeitsphysiologie an der Universitaet Dortmund Ardeystrasse 67, D-44139 Dortmund, Tel. 0231-1084-257 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide!
Re: [R] dataframes with only one variable
On Wed, 11 Jan 2006, Erich Neuwirth wrote: Subsetting from a dataframe with only one variable returns a vector, not a dataframe. This seems somewhat inconsistent. Not at all. It is entirely consistent with matrix-like indexing (the form you used). Wouldn't it be better if subsetting would respect the structure completely? It depends how you do it. [sel1,] parallels a matrix, and drops dimensions unless drop == FALSE is supplied. [sel1] returns a one-column df, and [[sel1]] returns a vector. It is just a question of choosing the appropriate tool. And any changes to this sort of thing (from the White book) would break a lot of careful code. v1-1:4 v2-4:1 df1-data.frame(v1) df2-data.frame(v1,v2) sel1-c(TRUE,TRUE,TRUE,TRUE) df1[sel1,] [1] 1 2 3 4 df2[sel1,] v1 v2 1 1 4 2 2 3 3 3 2 4 4 1 -- Erich Neuwirth Institute for Scientific Computing and Didactic Center for Computer Science University of Vienna phone: +43-1-4277-39464 fax: +43-1-4277-39459 -- 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
[R] Indentation in emacs
I'm using emacs-21.4 on debian unstable, together with the latest ESS implementation. I try to change indentation to 4 by following the advise in R-exts: It results in the following lines in my .emacs file: (custom-set-variables ;; custom-set-variables was added by Custom -- don't edit or cut/paste it! ;; Your init file should contain only one such instance. '(c-basic-offset 4) '(c-default-style bsd) '(latin1-display t nil (latin1-disp))) (custom-set-faces ;; custom-set-faces was added by Custom -- don't edit or cut/paste it! ;; Your init file should contain only one such instance. ) But it doesn't work with R code (with C code it works). So what is missing? -- Göran Broström __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Repeated measures aov with post hoc tests?
Dear list, I posted the message below a cople of days ago, and have not been able to find any solution to this. I do really want some help. /Fredrik On 1/10/06, Fredrik Karlsson [EMAIL PROTECTED] wrote: Dear list, I would like to perform an analysis on the following model: aov(ampratio ~ Type * Place * agemF + Error(speakerid/Place) ,data=aspvotwork) using the approach from http://www.psych.upenn.edu/~baron/rpsych/rpsych.html . Now, I got the test results, wich indicate a significant interaction and main effects of the agemF variable. How do I find at what level of agemF the effect may be found. How do I do this? I found a reference to TukeyHSD in the archives, but I cannot use it: TukeyHSD(aov(ampratio ~ Type * Place * agemF + Error(speakerid/Place),data=aspvotwork)) Error in TukeyHSD(aov(ampratio ~ Type * Place * agemF + Error(speakerid/Place), : no applicable method for TukeyHSD Please help me. /Fredrik -- My Gentoo + PVR-350 + IVTV + MythTV blog is on http://gentoomythtv.blogspot.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
Re: [R] a series of 1's and -1's
You could try to zip your data file and see whether there is a change in
file size that you feel is significant in which case the series is not
random (-:
To be sure, there is no such thing as a positive test for randomness, only
tests for specific deviations of randomness of which the runs.test and the
entropy are just two examples. Zip programs use a whole bunch of these tests
to compress files by finding structure in the data.
best, ingmar
From: Roger Bivand [EMAIL PROTECTED]
Reply-To: [EMAIL PROTECTED]
Date: Thu, 12 Jan 2006 08:42:25 +0100 (CET)
To: Mark Leeds [EMAIL PROTECTED]
Cc: R-Stat Help R-help@stat.math.ethz.ch
Subject: Re: [R] a series of 1's and -1's
On Wed, 11 Jan 2006, Mark Leeds wrote:
Does anyone know of a simple test
in any R package that given
a series of negative ones and positive
ones ( no other values are possible in the series )
returns a test of whether the series is random or not.
( a test at each point would be good but
I can use the apply function to implement
that ) ?
help.search(runs) points to function runs.test() in package tseries,
with examples:
x - factor(sign(rnorm(100))) # randomness
runs.test(x)
x - factor(rep(c(-1, 1), 50)) # over-mixing
runs.test(x)
which looks like your case
thanks.
**
This email and any files transmitted with it are confidentia...{{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
--
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: [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
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Equal number of cuts in a contourplot with trellis
Dear R-helpers, I need some help to produce a set of contour plots that I am trying to make in order to compare surfaces between the levels of a factor. For example: library(lattice) g - expand.grid(x = 60:100, y = 1:25, ti = c(a,b,c)) g$z -with(g, (-1e-4*x-1e-3*y-1e-5*x*y)*(ti==a) + (1e-2*x-1e-3*y-1e-4*x*y)*(ti==b) + (1e-3*x-1e-3*y-1.e-5*x*y)*(ti==c) ) contourplot(z~ x * y|ti, data = g, cuts=20, pretty=T, screen = list(z = 30, x = -60)) As you can see in the figure, most of the contour lines are in one of the levels and we are not able to see how the other levels look like. I would like to display the same number of cuts in each of the trellis. I can make each of the contourplots separately and control the number of cuts but I am not able to plot all of them in one. Thanks in advance, Jesus Jesús María Frías Celayeta School of Food Science and Environmental Health Dublin Institute of Technology Cathal Brugha St. Dublin 1 p + 353 1 4024459 f + 353 1 4024495 w http://www.dit.ie/DIT/tourismfood/science/ -- This message has been scanned for content and viruses by the DIT Information Services MailScanner Service, and is believed to be clean. http://www.dit.ie __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Software Reliability Using Cox Proportional Hazard
Hello,
i want to use coxph() for software reliability analysis, i use the
following implementation
###
failure-read.table(failure.dat, header=T)
attach(failure)
f-FailureNumber
t-TimeBetweenFailure
# filling vector f with ones
for(i in 1:length(f)) {
f[i]-1
}
library(survival)
# cox proportional hazard with covariable LOC
cox.res-coxph(Surv(t,f)~LOC,data=failure)
plot(survfit(cox.res))
##
failure.dat is
FailureNumber TimeBetweenFailureLOC
17120
211135
38141
410150
515162
622169
720170
825181
928188
1035193
here TimeBetweenFailure gives the time between following failures and
LOC is the actuall Lines of Code.
What i do, is filling the vector f with ones. f is then in Surv the
censoring variable. in survival analysis the censoring variable is 1 if
the patient dies, now i interpret the event death as
the appearence of a failure. And i use LOC as covariable.
My question now is, is this approach right or is there any big mistake
or wrong interpretation?
Thanks a Lot!
best regards
Andreas
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] a series of 1's and -1's
You could try to zip your data file and see whether there is a change in file size that you feel is significant in which case the series is not random (-: ... after converting the -1s and 1s to bits, of course. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Obtaining the adjusted r-square given the regression
Hi people, I want to obtain the adjusted r-square given a set of coefficients (without the intercept), and I don't know if there is a function that does it. Exist Dear Alexandra, Without knowing what routine you were using that returned an R-Square value too you, it is a little difficult to tell. I am not able to comment on your routine, but I'll try to give a satisfactory response to your first question. Assuming that you used lm() for your linear regression, then all you need to do is see the help page for the lm() summary method, which explains how to get at the values in the summary. In your console type the following: ?summary.lm That help page describes how to get at the values in the summary, and provides an example with getting the full correlation value (whereas the one printed is in a 'pretty' form). I assume the other ones can be accessed similarly, including the adjusted R^2. Your code might look something like the following: # Where object is the result of a call to lm() summary(object)$adj.r.squared The help also has a demonstration for how to exclude the intercept, where you simply add a -1 term too the end of your formula. There may still be an accessor function available, which I do not know about (BUT am certainly interested in if anyone has feedback). In the future, it would help people on the list decode your function when it too is printed in a readable form. The layout of your function in your message may have become garbled when it was stripped of all of its HTML information. If the code was formatted in a readable way when you sent the message, then that's what happened. If that is the case try composing your message in plain text to begin with. I hope this helps. Rgds, KeithC. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Space between axis label and tick labels
Hello Kare: I think some global graphics settings can get you what you need, assuming you are using base graphics functions such as plot(). See ?par for comprehensive details. For the specific characteristics you mention, I have found the arguments in par for: 1) font size of axis and tick labels 2) font thickness cex.axis cex.lab are helpful; and perhaps the font.* series 3) placement of both axis and tick labels mgp tck are helpful. Hope that helps, Bill --- Bill Pikounis, PhD Nonclinical Statistics Centocor, Inc. Malvern, PA 19355 (USA) -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Kare Edvardsen Sent: Wednesday, January 11, 2006 4:52 AM To: R-help Subject: [R] Space between axis label and tick labels I'm writing an publication in two column format and need to shrink some plots. After increasing the axis labels it does not look nice at all. The y-axis label and tick labels almost touch each other and the x-axis tick labels expand into the plot instead of away from it. Is there a better way than cex to control the: 1) font size of axis and tick labels 2) font thickness 3) placement of both axis and yick labels Cheers, Kare -- ### Kare Edvardsen [EMAIL PROTECTED] Norwegian Institute for Air Research (NILU) Polarmiljosenteret NO-9296 Tromso http://www.nilu.no Swb. +47 77 75 03 75 Dir. +47 77 75 03 90 Fax. +47 77 75 03 76 Mob. +47 90 74 60 69 ### __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Problem with NLSYSTEMFIT()
Hello, I want to solve a nonlinear 3SLS problem with nlsystemfit(). The equations are of the form y_it = f_i(x,t,theta) The functions f_i(.) have to be formulated as R-functions. When invoking nlsystemfit() I get the error Error in deriv.formula(eqns[[i]], names(parmnames)) : Function 'f1' is not in the derivatives table Isn't it possible to provide equations in the form eq1 ~ f1(x,t,theta) etc. to nlsystemfit() ? Kind regards, Kilian Plank [[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
[R] envelopes of simulations
Hello! I am writing you because I could not plot the confidence envelopes for functions Jest, Jcross, Jdot, Jmulti, and L, using the Spatstat package. I have already understood how to do that for Kest or Jest, that is: JEnv - plot(envelope(PPPData, Jest)) Where PPPData is my ppp object. However, for Jcross I must specify the two marks I want to analyse. That is, usually I would get the Jcross doing: Jc - Jcross(PPPData,Aun,Qsu) For marks Aun and Qsu. For L function, I can make: K - Kest (PPPData, correction=isotropic) plot (K, r-sqrt(iso/pi)~r) But I could not understand how to put all that together. Resuming, I do not understand how can I use the envelope function: - to plot Jcross (where and how do I specify types/marks ??) - to plot L (where do I write that I do not want K, but r-sqrt(iso/pi)~r, instead??) - etc. for similar functions not so simple as Jest, Kest or Gest. Hope someone can please help me on this. I have already looked at the package's manual, but I could not find any example similar to what I am trying to do. Best Regards, Sara Mouro. [[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
Re: [R] SPSS and R ? do they like each other?
Thank you very much! write.SPSS works fine.
I just wonder, why this very useful function is not part of any package. I have
not even found it in the web. For experts it may not be a big deal to write
their own export functions, but for newcomers like me it is almost impossible -
and at the same time it is essential to us to have good facilities for exchange
with our familiar statistical package. I think a lack of exchange tools might
be something that scares many people off and keeps them from getting to know R.
Well, just do give a greenhorn ´s perspective.
Best regards,
Michael
-Ursprüngliche Nachricht-
Von: Chuck Cleland [mailto:[EMAIL PROTECTED]
Gesendet: Donnerstag, 12. Januar 2006 01:16
An: Michael Reinecke
Cc: R-help@stat.math.ethz.ch
Betreff: Re: [R] SPSS and R ? do they like each other?
Michael Reinecke wrote:
Thanks again for your answer! I tried it out. write.foreign produces SPSS
syntax, but unfortunally this syntax tells SPSS to take the names (and not
the labels) in order to produce SPSS variable labels. The former labels get
lost.
I tried a data frame produced by read.spss and one by spss.get. Here is the
read.spss one (the labels meant to be exported are called Text 1, ...):
jjread- read.spss(test2.sav, use.value.labels=TRUE,
to.data.frame=TRUE)
str(jjread)
`data.frame': 30 obs. of 3 variables:
$ VAR1: num 101 102 103 104 105 106 107 108 109 110 ...
$ VAR2: num 6 6 5 6 6 6 6 6 6 6 ...
$ VAR3: num 0 0 6 7 0 7 0 0 0 8 ...
- attr(*, variable.labels)= Named chr Text 1 Text2 text 3
..- attr(*, names)= chr VAR1 VAR2 VAR3
datafile-tempfile()
codefile-tempfile()
write.foreign(jjread,datafile,codefile,package=SPSS)
file.show(datafile)
file.show(codefile)
The syntax file I get is:
DATA LIST FILE=
C:\DOKUME~1\reinecke\LOKALE~1\Temp\Rtmp15028\file27910 free / VAR1
VAR2 VAR3 .
VARIABLE LABELS
VAR1 VAR1
VAR2 VAR2
VAR3 VAR3
.
EXECUTE.
I am working on R 2.2.0. But I think a newer version won ´t fix it either,
will it?
Here is a functiong based on modifying foreign:::writeForeignSPSS (by Thomas
Lumley) which might work for you:
write.SPSS - function (df, datafile, codefile, varnames = NULL) { adQuote -
function(x){paste(\, x, \, sep = )}
dfn - lapply(df, function(x) if (is.factor(x))
as.numeric(x)
else x)
write.table(dfn, file = datafile, row = FALSE, col = FALSE)
if(is.null(attributes(df)$variable.labels)) varlabels - names(df) else
varlabels - attributes(df)$variable.labels
if (is.null(varnames)) {
varnames - abbreviate(names(df), 8)
if (any(sapply(varnames, nchar) 8))
stop(I cannot abbreviate the variable names to eight or fewer
letters)
if (any(varnames != names(df)))
warning(some variable names were abbreviated)
}
cat(DATA LIST FILE=, dQuote(datafile), free\n, file = codefile)
cat(/, varnames, .\n\n, file = codefile, append = TRUE)
cat(VARIABLE LABELS\n, file = codefile, append = TRUE)
cat(paste(varnames, adQuote(varlabels), \n), .\n, file = codefile,
append = TRUE)
factors - sapply(df, is.factor)
if (any(factors)) {
cat(\nVALUE LABELS\n, file = codefile, append = TRUE)
for (v in which(factors)) {
cat(/\n, file = codefile, append = TRUE)
cat(varnames[v], \n, file = codefile, append = TRUE)
levs - levels(df[[v]])
cat(paste(1:length(levs), adQuote(levs), \n, sep = ),
file = codefile, append = TRUE)
}
cat(.\n, file = codefile, append = TRUE)
}
cat(\nEXECUTE.\n, file = codefile, append = TRUE) }
--
Chuck Cleland, Ph.D.
NDRI, Inc.
71 West 23rd Street, 8th floor
New York, NY 10010
tel: (212) 845-4495 (Tu, Th)
tel: (732) 452-1424 (M, W, F)
fax: (917) 438-0894
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] extract variables from linear model
Hi, I fitted a linear model: fit - lm(y ~ a * b + c - 1 , na.action='na.omit') Now I want to extract only the a * b effects with confidence intervals. Of course, I can just add the coefficients by hand, but I think there should an easier way. I tried with predict.lm using the 'terms' argument, but I didn't manage to do it. Any hints are appreciated, best, joerg __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] SPSS and R ? do they like each other?
Michael Reinecke wrote:
Thank you very much! write.SPSS works fine.
I just wonder, why this very useful function is not part of any package. I
have not even found it in the web. For experts it may not be a big deal to
write their own export functions, but for newcomers like me it is almost
impossible - and at the same time it is essential to us to have good
facilities for exchange with our familiar statistical package. I think a lack
of exchange tools might be something that scares many people off and keeps
them from getting to know R.
Well, just do give a greenhorn ´s perspective.
...
The tool for exporting to SPSS *is* available in the foreign package
thanks to Thomas Lumley. I just made a *small modification* to use the
variable.labels attribute of a data frame if it's available and the
names of the data frame if that attribute is not available. Maybe
Thomas will consider making a change to foreign:::writeForeignSPSS along
those lines.
Chuck Cleland
-Ursprüngliche Nachricht-
Von: Chuck Cleland [mailto:[EMAIL PROTECTED]
Gesendet: Donnerstag, 12. Januar 2006 01:16
An: Michael Reinecke
Cc: R-help@stat.math.ethz.ch
Betreff: Re: [R] SPSS and R ? do they like each other?
Michael Reinecke wrote:
Thanks again for your answer! I tried it out. write.foreign produces SPSS
syntax, but unfortunally this syntax tells SPSS to take the names (and not
the labels) in order to produce SPSS variable labels. The former labels get
lost.
I tried a data frame produced by read.spss and one by spss.get. Here is the
read.spss one (the labels meant to be exported are called Text 1, ...):
jjread- read.spss(test2.sav, use.value.labels=TRUE,
to.data.frame=TRUE)
str(jjread)
`data.frame': 30 obs. of 3 variables:
$ VAR1: num 101 102 103 104 105 106 107 108 109 110 ...
$ VAR2: num 6 6 5 6 6 6 6 6 6 6 ...
$ VAR3: num 0 0 6 7 0 7 0 0 0 8 ...
- attr(*, variable.labels)= Named chr Text 1 Text2 text 3
..- attr(*, names)= chr VAR1 VAR2 VAR3
datafile-tempfile()
codefile-tempfile()
write.foreign(jjread,datafile,codefile,package=SPSS)
file.show(datafile)
file.show(codefile)
The syntax file I get is:
DATA LIST FILE=
C:\DOKUME~1\reinecke\LOKALE~1\Temp\Rtmp15028\file27910 free / VAR1
VAR2 VAR3 .
VARIABLE LABELS
VAR1 VAR1
VAR2 VAR2
VAR3 VAR3
.
EXECUTE.
I am working on R 2.2.0. But I think a newer version won ´t fix it either,
will it?
Here is a functiong based on modifying foreign:::writeForeignSPSS (by Thomas
Lumley) which might work for you:
write.SPSS - function (df, datafile, codefile, varnames = NULL) { adQuote -
function(x){paste(\, x, \, sep = )}
dfn - lapply(df, function(x) if (is.factor(x))
as.numeric(x)
else x)
write.table(dfn, file = datafile, row = FALSE, col = FALSE)
if(is.null(attributes(df)$variable.labels)) varlabels - names(df) else
varlabels - attributes(df)$variable.labels
if (is.null(varnames)) {
varnames - abbreviate(names(df), 8)
if (any(sapply(varnames, nchar) 8))
stop(I cannot abbreviate the variable names to eight or fewer
letters)
if (any(varnames != names(df)))
warning(some variable names were abbreviated)
}
cat(DATA LIST FILE=, dQuote(datafile), free\n, file = codefile)
cat(/, varnames, .\n\n, file = codefile, append = TRUE)
cat(VARIABLE LABELS\n, file = codefile, append = TRUE)
cat(paste(varnames, adQuote(varlabels), \n), .\n, file = codefile,
append = TRUE)
factors - sapply(df, is.factor)
if (any(factors)) {
cat(\nVALUE LABELS\n, file = codefile, append = TRUE)
for (v in which(factors)) {
cat(/\n, file = codefile, append = TRUE)
cat(varnames[v], \n, file = codefile, append = TRUE)
levs - levels(df[[v]])
cat(paste(1:length(levs), adQuote(levs), \n, sep = ),
file = codefile, append = TRUE)
}
cat(.\n, file = codefile, append = TRUE)
}
cat(\nEXECUTE.\n, file = codefile, append = TRUE) }
--
Chuck Cleland, Ph.D.
NDRI, Inc.
71 West 23rd Street, 8th floor
New York, NY 10010
tel: (212) 845-4495 (Tu, Th)
tel: (732) 452-1424 (M, W, F)
fax: (917) 438-0894
__
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
--
Chuck Cleland, Ph.D.
NDRI, Inc.
71 West 23rd Street, 8th floor
New York, NY 10010
tel: (212) 845-4495 (Tu, Th)
tel: (732) 452-1424 (M, W, F)
fax: (917) 438-0894
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] dataframes with only one variable
df1 v1 1 1 2 2 3 3 4 4 df1[,] [1] 1 2 3 4 df1[,1] [1] 1 2 3 4 df1[,,drop=F] v1 1 1 2 2 3 3 4 4 df1[,1,drop=F] v1 1 1 2 2 3 3 4 4 df1[1] v1 1 1 2 2 3 3 4 4 df1[[1]] [1] 1 2 3 4 For transfers from Excel to R using the [put/get] R dataframe commands, I think it is important always to use the drop=FALSE argument (as I assume you are doing in RExcel V1.55). The reason for this is to maintain a rigid relationship between the only partially compatible conventions of Excel and R. For strictly within R use, the case is less clear. I have trained myself always (well 85% on the first try) to use the drop=FALSE argument when I care about the structure after the copy. The tension between keeping the structure and demoting the structure predates data.frames. This was a design issue in matrices as well. tmp - matrix(1:6,2,3) tmp [,1] [,2] [,3] [1,]135 [2,]246 tmp[1,] [1] 1 3 5 tmp[1,,drop=FALSE] [,1] [,2] [,3] [1,]135 Rich __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] SPSS and R ? do they like each other?
2006/1/12, Chuck Cleland [EMAIL PROTECTED]:
Michael Reinecke wrote:
Thank you very much! write.SPSS works fine.
I just wonder, why this very useful function is not part of any package. I
have not even found it in the web. For experts it may not be a big deal to
write their own export functions, but for newcomers like me it is almost
impossible - and at the same time it is essential to us to have good
facilities for exchange with our familiar statistical package. I think a
lack of exchange tools might be something that scares many people off and
keeps them from getting to know R.
Well, just do give a greenhorn ´s perspective.
...
The tool for exporting to SPSS *is* available in the foreign package
thanks to Thomas Lumley. I just made a *small modification* to use the
variable.labels attribute of a data frame if it's available and the
names of the data frame if that attribute is not available. Maybe
Thomas will consider making a change to foreign:::writeForeignSPSS along
those lines.
I agree with this point. it 's usefull when one get the spss data file
into R to do something and export that data back to spss data file.
Chuck Cleland
-Ursprüngliche Nachricht-
Von: Chuck Cleland [mailto:[EMAIL PROTECTED]
Gesendet: Donnerstag, 12. Januar 2006 01:16
An: Michael Reinecke
Cc: R-help@stat.math.ethz.ch
Betreff: Re: [R] SPSS and R ? do they like each other?
Michael Reinecke wrote:
Thanks again for your answer! I tried it out. write.foreign produces SPSS
syntax, but unfortunally this syntax tells SPSS to take the names (and not
the labels) in order to produce SPSS variable labels. The former labels get
lost.
I tried a data frame produced by read.spss and one by spss.get. Here is the
read.spss one (the labels meant to be exported are called Text 1, ...):
jjread- read.spss(test2.sav, use.value.labels=TRUE,
to.data.frame=TRUE)
str(jjread)
`data.frame': 30 obs. of 3 variables:
$ VAR1: num 101 102 103 104 105 106 107 108 109 110 ...
$ VAR2: num 6 6 5 6 6 6 6 6 6 6 ...
$ VAR3: num 0 0 6 7 0 7 0 0 0 8 ...
- attr(*, variable.labels)= Named chr Text 1 Text2 text 3
..- attr(*, names)= chr VAR1 VAR2 VAR3
datafile-tempfile()
codefile-tempfile()
write.foreign(jjread,datafile,codefile,package=SPSS)
file.show(datafile)
file.show(codefile)
The syntax file I get is:
DATA LIST FILE=
C:\DOKUME~1\reinecke\LOKALE~1\Temp\Rtmp15028\file27910 free / VAR1
VAR2 VAR3 .
VARIABLE LABELS
VAR1 VAR1
VAR2 VAR2
VAR3 VAR3
.
EXECUTE.
I am working on R 2.2.0. But I think a newer version won ´t fix it either,
will it?
Here is a functiong based on modifying foreign:::writeForeignSPSS (by
Thomas Lumley) which might work for you:
write.SPSS - function (df, datafile, codefile, varnames = NULL) { adQuote
- function(x){paste(\, x, \, sep = )}
dfn - lapply(df, function(x) if (is.factor(x))
as.numeric(x)
else x)
write.table(dfn, file = datafile, row = FALSE, col = FALSE)
if(is.null(attributes(df)$variable.labels)) varlabels - names(df)
else varlabels - attributes(df)$variable.labels
if (is.null(varnames)) {
varnames - abbreviate(names(df), 8)
if (any(sapply(varnames, nchar) 8))
stop(I cannot abbreviate the variable names to eight or fewer
letters)
if (any(varnames != names(df)))
warning(some variable names were abbreviated)
}
cat(DATA LIST FILE=, dQuote(datafile), free\n, file = codefile)
cat(/, varnames, .\n\n, file = codefile, append = TRUE)
cat(VARIABLE LABELS\n, file = codefile, append = TRUE)
cat(paste(varnames, adQuote(varlabels), \n), .\n, file = codefile,
append = TRUE)
factors - sapply(df, is.factor)
if (any(factors)) {
cat(\nVALUE LABELS\n, file = codefile, append = TRUE)
for (v in which(factors)) {
cat(/\n, file = codefile, append = TRUE)
cat(varnames[v], \n, file = codefile, append = TRUE)
levs - levels(df[[v]])
cat(paste(1:length(levs), adQuote(levs), \n, sep = ),
file = codefile, append = TRUE)
}
cat(.\n, file = codefile, append = TRUE)
}
cat(\nEXECUTE.\n, file = codefile, append = TRUE) }
--
Chuck Cleland, Ph.D.
NDRI, Inc.
71 West 23rd Street, 8th floor
New York, NY 10010
tel: (212) 845-4495 (Tu, Th)
tel: (732) 452-1424 (M, W, F)
fax: (917) 438-0894
__
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
--
Chuck Cleland, Ph.D.
NDRI, Inc.
71 West 23rd Street, 8th floor
New York, NY 10010
tel:
Re: [R] Indentation in emacs
On 12 Jan 2006, [EMAIL PROTECTED] wrote: I'm using emacs-21.4 on debian unstable, together with the latest ESS implementation. I try to change indentation to 4 by following the advise in R-exts: It results in the following lines in my .emacs file: (custom-set-variables ;; custom-set-variables was added by Custom -- don't edit or ;; cut/paste it! Your init file should contain only one such ;; instance. '(c-basic-offset 4) '(c-default-style bsd) '(latin1-display t nil (latin1-disp))) (custom-set-faces ;; custom-set-faces was added by Custom -- don't edit or cut/paste it! ;; Your init file should contain only one such instance. ) But it doesn't work with R code (with C code it works). So what is missing? Try: (setq ess-indent-level 4) You may also be interested in the ESS mail list (a better place for such questions): [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/ess-help + seth __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] tapply and weighted means
I' m trying to compute weighted mean on different groups but it only returns NA. If I use the following data.frame truc: x y w 1 1 1 1 2 2 1 3 1 1 4 2 0 2 1 0 3 2 0 4 1 0 5 1 where x is a factor, and then use the command : tapply(truc$y,list(truc$x),wtd.mean, weights=truc$w) I just get NA. What's the problem ? What can I do ? __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] tapply and weighted means
you need also to split the 'w' column, for each level of 'x'; you could use: lapply(split(truc, truc$x), function(z) weighted.mean(z$y, z$w)) 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://www.med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: Florent Bresson [EMAIL PROTECTED] To: R-help r-help@stat.math.ethz.ch Sent: Thursday, January 12, 2006 3:44 PM Subject: [R] tapply and weighted means I' m trying to compute weighted mean on different groups but it only returns NA. If I use the following data.frame truc: x y w 1 1 1 1 2 2 1 3 1 1 4 2 0 2 1 0 3 2 0 4 1 0 5 1 where x is a factor, and then use the command : tapply(truc$y,list(truc$x),wtd.mean, weights=truc$w) I just get NA. What's the problem ? What can I do ? __ 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 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
Re: [R] Problem with NLSYSTEMFIT()
Dear Kilian, On Thursday 12 January 2006 14:20, Kilian Plank wrote: Hello, I want to solve a nonlinear 3SLS problem with nlsystemfit(). The equations are of the form y_it = f_i(x,t,theta) The functions f_i(.) have to be formulated as R-functions. When invoking nlsystemfit() I get the error Error in deriv.formula(eqns[[i]], names(parmnames)) : Function 'f1' is not in the derivatives table Isn't it possible to provide equations in the form eq1 ~ f1(x,t,theta) etc. to nlsystemfit() ? Unfortunately, this is not (yet) possible. You have to specify your equations like eq1 - y ~ b0 + x1^b1 + b1^2 * log( x2) (see the documentation of nlsystemfit). Furthermore, I suggest that you ask the author of nlsystemfit (Jeff Hamann, see cc) how reliable the algorithms for the non-linear estimation are in the current version of systemfit (while the code for the linear estimation is very mature now, the non-linear estimation is still under development). You are invited to help us improving the code of nlsystemfit ;-) Best wishes, Arne Kind regards, Kilian Plank [[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 -- Arne Henningsen Department of Agricultural Economics University of Kiel Olshausenstr. 40 D-24098 Kiel (Germany) Tel: +49-431-880 4445 Fax: +49-431-880 1397 [EMAIL PROTECTED] http://www.uni-kiel.de/agrarpol/ahenningsen/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] tapply and weighted means
Dimitris Rizopoulos wrote: you need also to split the 'w' column, for each level of 'x'; you could use: lapply(split(truc, truc$x), function(z) weighted.mean(z$y, z$w)) I hope it helps. Best, Dimitris Or: library(Hmisc) ?wtd.mean The help file has a built-in example of this. Frank 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://www.med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: Florent Bresson [EMAIL PROTECTED] To: R-help r-help@stat.math.ethz.ch Sent: Thursday, January 12, 2006 3:44 PM Subject: [R] tapply and weighted means I' m trying to compute weighted mean on different groups but it only returns NA. If I use the following data.frame truc: x y w 1 1 1 1 2 2 1 3 1 1 4 2 0 2 1 0 3 2 0 4 1 0 5 1 where x is a factor, and then use the command : tapply(truc$y,list(truc$x),wtd.mean, weights=truc$w) I just get NA. What's the problem ? What can I do ? __ 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 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 -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] wilcox.test warnig message p-value where are the zeros in the data?
does anybody know why there are the two warnings in the example above? Regards Knut day_4 [1] 540 1 1 1 1 1 1 300 720 480 day_1 [1] 438 3431 4751 562 500 435 1045 890 is.vector (day_1) [1] TRUE is.vector (day_4) [1] TRUE wilcox.test(day_4 ,day_1,paired=TRUE,alternative=two.sided,exact=TRUE,conf.int=TRUE) Wilcoxon signed rank test with continuity correction data: day_4 and day_1 V = 1, p-value = 0.02086 alternative hypothesis: true mu is not equal to 0 95 percent confidence interval: -486.5 -120.0 sample estimates: (pseudo)median -348 Warning messages: 1: cannot compute exact p-value with zeroes in: wilcox.test.default(day_4, day_1, paired = TRUE, alternative = two.sided, 2: cannot compute exact confidence interval with zeroes in: wilcox.test.default(day_4, day_1, paired = TRUE, alternative = two.sided, __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] extract variables from linear model
I fitted a linear model: fit - lm(y ~ a * b + c - 1 , na.action='na.omit') wouldn't a simple coef(fit)[2] work? Jukka Ruohonen. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] wilcox.test warnig message p-value where are the zeros in the data?
Knut Krueger [EMAIL PROTECTED] writes: does anybody know why there are the two warnings in the example above? Regards Knut day_4 [1] 540 1 1 1 1 1 1 300 720 480 day_1 [1] 438 3431 4751 562 500 435 1045 890 is.vector (day_1) [1] TRUE is.vector (day_4) [1] TRUE The paired Wilcoxon test depends on pairwise differences and as far as I can see you have two of those being zero, in cases 3 and 5. wilcox.test(day_4 ,day_1,paired=TRUE,alternative=two.sided,exact=TRUE,conf.int=TRUE) Wilcoxon signed rank test with continuity correction data: day_4 and day_1 V = 1, p-value = 0.02086 alternative hypothesis: true mu is not equal to 0 95 percent confidence interval: -486.5 -120.0 sample estimates: (pseudo)median -348 Warning messages: 1: cannot compute exact p-value with zeroes in: wilcox.test.default(day_4, day_1, paired = TRUE, alternative = two.sided, 2: cannot compute exact confidence interval with zeroes in: wilcox.test.default(day_4, day_1, paired = TRUE, alternative = two.sided, __ 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 -- 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
Re: [R] tapply and weighted means
On Thu, 2006-01-12 at 15:44 +0100, Florent Bresson wrote: I' m trying to compute weighted mean on different groups but it only returns NA. If I use the following data.frame truc: x y w 1 1 1 1 2 2 1 3 1 1 4 2 0 2 1 0 3 2 0 4 1 0 5 1 where x is a factor, and then use the command : tapply(truc$y,list(truc$x),wtd.mean, weights=truc$w) I just get NA. What's the problem ? What can I do ? Florent, I guess you didn't read the help for tapply, which in the Value section states: Note that optional arguments to 'FUN' supplied by the '...' argument are not divided into cells. It is therefore inappropriate for 'FUN' to expect additional arguments with the same length as 'X'. So tapply is not the right tool for this job. We can use by() instead (a wrapper for tapply) as so: dat - matrix(scan(), byrow = TRUE, ncol = 3) 1 1 1 1 2 2 1 3 1 1 4 2 0 2 1 0 3 2 0 4 1 0 5 1 colnames(dat) - c(x, y, w) dat - as.data.frame(dat) dat (res - by(dat, dat$x, function(z) weighted.mean(z$y, z$w))) but if you want to easily access the numbers you need to do a little work, e.g. as.vector(res) Also, I don't see a function wtd.mean in standard R and weighted.mean() doesn't have a weights argument, so I guess you are using a function from another package and did not tell us. HTH, Gav -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Gavin Simpson [T] +44 (0)20 7679 5522 ENSIS Research Fellow [F] +44 (0)20 7679 7565 ENSIS Ltd. ECRC [E] gavin.simpsonATNOSPAMucl.ac.uk UCL Department of Geography [W] http://www.ucl.ac.uk/~ucfagls/cv/ 26 Bedford Way[W] http://www.ucl.ac.uk/~ucfagls/ London. WC1H 0AP. %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] I think simple R question
I have a vector x with #'s ( 1 or -1 in them ) in it and I want to
mark a new vector with the sign of the value of the a streak
of H where H = some number ( at the next spot in the vector )
So, say H was equal to 3 and
I had a vector of
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
1 -11 11 -11 1 -1-1
then, I would want a function to return a new
vector of
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
0 000 0 1 0 0 0 0
As I said, I used to do these things like this
it's been a while and I'm rusty with this stuff.
Without looping is preferred but looping is okay
also.
Mark
**
This email and any files transmitted with it are confidentia...{{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
Re: [R] I think simple R question
see ?rle
url:www.econ.uiuc.edu/~rogerRoger Koenker
email[EMAIL PROTECTED]Department of Economics
vox: 217-333-4558University of Illinois
fax: 217-244-6678Champaign, IL 61820
On Jan 12, 2006, at 9:56 AM, Mark Leeds wrote:
I have a vector x with #'s ( 1 or -1 in them ) in it and I want to
mark a new vector with the sign of the value of the a streak
of H where H = some number ( at the next spot in the vector )
So, say H was equal to 3 and
I had a vector of
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
1 -11 11 -11 1 -1-1
then, I would want a function to return a new
vector of
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
0 000 0 1 0 0 0 0
As I said, I used to do these things like this
it's been a while and I'm rusty with this stuff.
Without looping is preferred but looping is okay
also.
Mark
**
This email and any files transmitted with it are confidentia...
{{dropped}}
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-
guide.html
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] infinite recursion in do.call when lme4 loaded only
A larg program which worked with lme4/R about a year ago failed when I
re-run it today. I reproduced the problem with the program below.
-- When lme4 is not loaded, the program runs ok and fast enough
-- When lme4 is loaded (but never used), the do.call fails
with infinite recursion after 60 seconds. Memory used increases
beyond bonds in task manager.
-- I tested a few S3 based packages (MASS, nlme) and did not get
similar problems
Current workaround: do lme4-processing in a separate program.
--
#library(lme4) # uncomment this to see the problem
np - 12
nq - 20
nph - 3
nrep - 30
grd - expand.grid(Pat=as.factor(1:np),
Q=as.factor(1:nq),
Phase=as.factor(1:nph))
df - with (grd,
data.frame(Pat=Pat,Q=Q,Phase=Phase,Resp = rnorm(np*nq*nph*nrep)))
score - function(x) {
data.frame(Pat=x$Pat[1],Phase=x$Phase[1],Q=x$Q[1],score = mean(x$Resp))
}
# about 20 sec
caScore - by(df,list(Pat=df$Pat,Phase=df$Phase,Q=df$Q),FUN = score )
ca1 = do.call(rbind,caScore)
# Without lme4: 3 seconds
# With lme4: After 60 sec:
#Error: evaluation nested too deeply: infinite recursion /
#options(expressions=)?
---
platform i386-pc-mingw32
arch i386
os mingw32
system i386, mingw32
status
major2
minor2.1
year 2005
month12
day 20
svn rev 36812
language R
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Equal number of cuts in a contourplot with trellis
On 1/12/06, Jesus Frias [EMAIL PROTECTED] wrote:
Dear R-helpers,
I need some help to produce a set of contour plots that I am
trying to make in order to compare surfaces between the levels of a
factor. For example:
library(lattice)
g - expand.grid(x = 60:100, y = 1:25, ti = c(a,b,c))
g$z -with(g,
(-1e-4*x-1e-3*y-1e-5*x*y)*(ti==a) +
(1e-2*x-1e-3*y-1e-4*x*y)*(ti==b) +
(1e-3*x-1e-3*y-1.e-5*x*y)*(ti==c)
)
contourplot(z~ x * y|ti, data = g,
cuts=20,
pretty=T,
screen = list(z = 30, x = -60))
As you can see in the figure, most of the contour lines are in one of
the levels and we are not able to see how the other levels look like.
I would like to display the same number of cuts in each of the trellis.
I can make each of the contourplots separately and control the number of
cuts but I am not able to plot all of them in one.
The simplest solution is to recompute the levels for each panel function:
contourplot(z~ x * y|ti, data = g,
label.style = align,
panel = function(x, y, z, subscripts, at, ...) {
at - pretty(z[subscripts], 10)
panel.contourplot(x, y, z,
subscripts = subscripts,
at = at,
...)
})
Alternatively, you could pass in a suitable 'at' vector computed externally.
Deepayan
--
http://www.stat.wisc.edu/~deepayan/
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Repeated measures aov with post hoc tests?
The multcomp package may do what you want (there is mention of nested variables in the help). -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare [EMAIL PROTECTED] (801) 408-8111 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Fredrik Karlsson Sent: Thursday, January 12, 2006 3:24 AM To: r-help@stat.math.ethz.ch Subject: Re: [R] Repeated measures aov with post hoc tests? Dear list, I posted the message below a cople of days ago, and have not been able to find any solution to this. I do really want some help. /Fredrik On 1/10/06, Fredrik Karlsson [EMAIL PROTECTED] wrote: Dear list, I would like to perform an analysis on the following model: aov(ampratio ~ Type * Place * agemF + Error(speakerid/Place) ,data=aspvotwork) using the approach from http://www.psych.upenn.edu/~baron/rpsych/rpsych.html . Now, I got the test results, wich indicate a significant interaction and main effects of the agemF variable. How do I find at what level of agemF the effect may be found. How do I do this? I found a reference to TukeyHSD in the archives, but I cannot use it: TukeyHSD(aov(ampratio ~ Type * Place * agemF + Error(speakerid/Place),data=aspvotwork)) Error in TukeyHSD(aov(ampratio ~ Type * Place * agemF + Error(speakerid/Place), : no applicable method for TukeyHSD Please help me. /Fredrik -- My Gentoo + PVR-350 + IVTV + MythTV blog is on http://gentoomythtv.blogspot.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 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] infinite recursion in do.call when lme4 loaded only
Dieter Menne [EMAIL PROTECTED] writes:
A larg program which worked with lme4/R about a year ago failed when I
re-run it today. I reproduced the problem with the program below.
-- When lme4 is not loaded, the program runs ok and fast enough
-- When lme4 is loaded (but never used), the do.call fails
with infinite recursion after 60 seconds. Memory used increases
beyond bonds in task manager.
-- I tested a few S3 based packages (MASS, nlme) and did not get
similar problems
Current workaround: do lme4-processing in a separate program.
Looks like it conks out when the number of frames to rbind is bigger
than about 110.
Current releases have
options(expressions)
$expressions
[1] 1000
It was 5000 for a while, but we found that it could overflow the C
stack on some systems. Since your example has 720 lines, I can't quite
rule out that that the problem was really there all the time.
However, it surely has to do with methods dispatch:
system.time(do.call(rbind.data.frame,caScore))
[1] 0.99 0.00 0.99 0.00 0.00
which provides you with another workaround.
--
#library(lme4) # uncomment this to see the problem
np - 12
nq - 20
nph - 3
nrep - 30
grd - expand.grid(Pat=as.factor(1:np),
Q=as.factor(1:nq),
Phase=as.factor(1:nph))
df - with (grd,
data.frame(Pat=Pat,Q=Q,Phase=Phase,Resp = rnorm(np*nq*nph*nrep)))
score - function(x) {
data.frame(Pat=x$Pat[1],Phase=x$Phase[1],Q=x$Q[1],score = mean(x$Resp))
}
# about 20 sec
caScore - by(df,list(Pat=df$Pat,Phase=df$Phase,Q=df$Q),FUN = score )
ca1 = do.call(rbind,caScore)
# Without lme4: 3 seconds
# With lme4: After 60 sec:
#Error: evaluation nested too deeply: infinite recursion /
#options(expressions=)?
---
platform i386-pc-mingw32
arch i386
os mingw32
system i386, mingw32
status
major2
minor2.1
year 2005
month12
day 20
svn rev 36812
language R
__
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
--
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
[R] extract variables from linear model
Hi, I fitted a mixed linear model y = a + b + a*b + c + error, with c being the random factor: lmefit - lme(y ~ a * b - 1 , random = ~ 1 | c, na.action='na.omit') Is there a way to omit some level combinations of the cross-term a:b? E.g. those that are not significant? When I add the coefficients of a, b and a:b to get the combined effect without the effect of c, do then the standard errors of the coefficients also add? best, joerg __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R - Wikis and R-core
Hello Martin and others, I am happy with this decision. I'll look a little bit at this next week. Best, Philippe Grosjean We've had a small review time within R-core on this topic, amd would like to state the following: -- The R-core team welcomes proposals to develop an R-wiki. - We would consider linking a very small number of Wikis (ideally one) from www.r-project.org and offering an address in the r-project.org domain (such as 'wiki.r-project.org'). - The core team has no support time to offer, and would be looking for a medium-term commitment from a maintainer team for the Wiki(s). - Suggestions for the R documentation would best be filtered through the Wiki maintainers, who could e.g. supply suggested patches during the alpha phase of an R release. -- Our main concerns have been about ensuring the quality of such extra documentation projects, hence the 2nd point above. Several of our more general, not mainly R, experiences have been of outdated web pages which are continued to be used as reference when their advice has long been superseded. I think it's very important to try ensuring that this won't happen with an R Wiki. Martin Maechler, ETH Zurich PhGr == Philippe Grosjean [EMAIL PROTECTED] on Sun, 8 Jan 2006 17:00:44 +0100 (CET) writes: PhGr Hello all, Sorry for not taking part of this PhGr discussion earlier, and for not answering Detlef PhGr Steuer, Martin Maechler, and others that asked more PhGr direct questions to me. I am away from my office and PhGr my computer until the 16th of January. PhGr Just quick and partial answers: 1) I did not know PhGr about Hamburg RWiki. But I would be happy to merge PhGr both in one or the other way, as Detlef suggests it. PhGr 2) I choose DokuWiki as the best engine after a PhGr careful comparison of various Wiki engines. It is the PhGr best one, as far as I know, for the purpose of PhGr writting software documentation and similar PhGr pages. There is an extensive and clearly presented PhGr comparison of many Wikki engines at: PhGr http://www.wikimatrix.org/. PhGr 3) I started to change DokuWiki (addition of various PhGr plugins, addition of R code syntax coloring with PhGr GESHI, etc...). So, it goes well beyond all current PhGr Wiki engines regarding its suitability to present R PhGr stuff. PhGr 4) The reasons I did this is because I think the Wiki PhGr format could be of a wider use. I plan to change a PhGr little bit the DokuWiki syntax, so that it works with PhGr plain .R code files (Wiki part is simply embedded in PhGr commented lines, and the rest is recognized and PhGr formatted as R code by the Wiki engine). That way, the PhGr same Wiki document can either rendered by the Wiki PhGr engine for a nice presentation, or sourced in R PhGr indifferently. PhGr 5) My last idea is to add a Rpad engine to the Wiki, PhGr so that one could play with R code presented in the PhGr Wiki pages and see the effects of changes directly in PhGr the Wiki. PhGr 6) Regarding the content of the Wiki, it should be PhGr nice to propose to the authors of various existing PhGr document to put them in a Wiki form. Something like PhGr Statistics with R PhGr (http://zoonek2.free.fr/UNIX/48_R/all.html) is written PhGr in a way that stimulates additions to pages in PhGr perpetual construction, if it was presented in a Wiki PhGr form. It is licensed as Creative Commons PhGr Attribution-NonCommercial-ShareAlike 2.5 license, that PhGr is, exactly the same one as DokuWiki that I choose for PhGr R Wiki. Of course, I plan to ask its author to do so PhGr before putting its hundreds of very interesting pages PhGr on the Wiki... I think it is vital to have already PhGr something in the Wiki, in order to attract enough PhGr readers, and then enough contributors! PhGr 7) Regarding spamming and vandalism, DokuWiki allows PhGr to manage rights and users, even individually for PhGr pages. I think it would be fine to lock pages that PhGr reach a certain maturity (read-only / editable by PhGr selected users only) , with link to a discussion page PhGr which remaining freely accessible at the bottom of PhGr locked pages. PhGr 8) I would be happy to contribute this work to the R PhGr foundation in one way or the other to integrate it in PhGr http://www.r-project.org or PhGr http://cran.r-project.org. But if it is fine keeping PhGr it in http://www.sciviews.org as well, it is also fine PhGr for me. PhGr I suggest that all interested people drop a little PhGr email to my mailbox. I'll recontact
[R] Curve fitting
Hi! I have a problem of curve fitting. I use the following data : - vector of predictor data : 0 0.4 0.8 1.2 1.6 - vector of response data : 0.81954 0.64592 0.51247 0.42831 0.35371 I perform parametric fits using custom equations when I use this equation : y = yo + K *(1/(1+exp(-(a+b*ln(x) the fitting result is OK but when I use this more general equation :y = yo + K *(1/(1+exp(-(a+b*log(x)+c*x , then I get an aberrant curve! I don't understand that... The second fitting should be at least as good as the first one because when taking c=0, both equations are identical! There is here a mathematical phenomenon that I don't understand!could someone help me Thanks a lot in advance! Nadège [[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
Re: [R] infinite recursion in do.call when lme4 loaded only
Peter Dalgaard p.dalgaard at biostat.ku.dk writes: A larg program which worked with lme4/R about a year ago failed when I re-run it today. I reproduced the problem with the program below. -- When lme4 is loaded (but never used), the do.call fails with infinite recursion after 60 seconds. Memory used increases beyond bonds in task manager. However, it surely has to do with methods dispatch: system.time(do.call(rbind.data.frame,caScore)) [1] 0.99 0.00 0.99 0.00 0.00 which provides you with another workaround. Peter, I had increased the optional value already, but I still don't understand what this recursion overflow has to do with the lm4 loading. Dieter __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] t-test for standard deviations
Dear R-users, I am new to the list and I would like to submit (probably) a stupid question: I found in a paper a reference to a t-test for the evaluationg the difference between the standard deviations of 2 samples. This test is performed in the paper but the methodology is not explained and any reference is reported. Does anyone know where I can find references to this test and if it is implemented in R? Thenks in advance for your help, Mirko [[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
[R] CRAN versions of lme4/Matrix don't appear to work with R 2.1.1
I am currently using R 2.1.1 under Windows and I do not seem to be able to load the current versions of lme4/Matrix. I have run 'update.packages.' I understand this is still experimental software but I would like access to a working version. Thanks. Chuck R Output: library(lme4) Loading required package: Matrix Error in lazyLoadDBfetch(key, datafile, compressed, envhook) : ReadItem: unknown type 241 In addition: Warning messages: 1: package 'lme4' was built under R version 2.3.0 2: package 'Matrix' was built under R version 2.3.0 Error: package 'Matrix' could not be loaded __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] t-test for standard deviations
On 12-Jan-06 mirko sanpietrucci wrote: Dear R-users, I am new to the list and I would like to submit (probably) a stupid question: I found in a paper a reference to a t-test for the evaluationg the difference between the standard deviations of 2 samples. This test is performed in the paper but the methodology is not explained and any reference is reported. Does anyone know where I can find references to this test and if it is implemented in R? Thenks in advance for your help, Mirko If the paper says that a 1) t-test was used for evaluating the difference between the 2) standard deviations of 2 samples then I suspect that one or the other of these is a misprint. To compare standard deviations (more precisely, variances) you could use a (1)F-test. Or you would use a t-test to evaluate the difference between the (2)means of 2 samples. If it is really obscure what was done, perhaps an appropriate quotation from the paper would help to ascertain the problem. Best wishes, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 12-Jan-06 Time: 18:52:31 -- XFMail -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] I think not so hard question
Mark,
It's a bit unclear whether you're looking for a run of exactly 3, or at
least 3. If it's exactly 3, what Prof. Koenker suggested (using rle())
should help you a lot. If you want runs of at least 3, it should work
similarly as well.
set.seed(1)
(x - sample(c(-1, 1), 100, replace=TRUE))
[1] -1 -1 1 1 -1 1 1 1 1 -1 -1 -1 1 -1 1 -1 1 1 -1 1 1 -1 1
-1 -1 -1
[27] -1 -1 1 -1 -1 1 -1 -1 1 1 1 -1 1 -1 1 1 1 1 1 1 -1 -1 1
1 -1 1
[53] -1 -1 -1 -1 -1 1 1 -1 1 -1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 -1
1 1 -1
[79] 1 1 -1 1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 1 1 1 1 -1 -1 1 1
r - rle(x)$length
p - which(r == 3)
idx - cumsum(r)[p] - 2
sapply(idx, function(i) x[i:(i+2)])
[,1] [,2] [,3] [,4]
[1,] -11 -1 -1
[2,] -11 -1 -1
[3,] -11 -1 -1
idx
[1] 10 35 62 73
Andy
From: Mark Leeds
I'm sorry to bother this list so much
But I haven't programmed in
A while and I'm struggling.
I have a vector in R of 1's and -1's
And I want to use a streak of size Y
To predict that the same value will
Be next.
So, suppose Y = 3. Then, if there is a streak of three
ones in a row, then I will predict that the next value is
a 1. But, if there is a streak of 3 -1's in a row,
then I will predict that a -1 is next. Otherwise,
I don't predict anything.
I am really new to R and kind of struggling
And I was wondering if someone could show
how to do this ?
In other words, given a vector of -1's
And 1's, I am unable ( I've been trying
For 2 days ) to create a new vector that
Has the predictions in it at the appropriate
places ? Thanks.
**
This email and any files transmitted with it are
confidentia...{{dropped}}
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Multilevel models with mixed effects in R?
Group,
I am new to R. In my work as a program evaluator, I am regularly asked
to estimate effect sizes of prevention/intervention and educational
programs on various student outcomes (e.g. academic achievement). In
many cases, I have access to data over three or more time periods (e.g.
growth in proficiency test scores).
I usually have multiple independent and dependent variables in each
model along with covariates. I have historically utilized latent growth
curve structural equation models, but would like to include random
effects in the model. Does R have the ability to run such analyses?
Regards,
Charles R. Partridge
Evaluation Specialist
Center for Learning Excellence
The John Glenn Institute for Public Service Public Policy
807 Kinnear Road, Room 214
Columbus, Ohio 43212-1421
Phone: 614.292.2419
FAX: 614.247.6447
Email: [EMAIL PROTECTED]
http://cle.osu.edu
CONFIDENTIALITY NOTICE: This message is intended only for th...{{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
Re: [R] Curve fitting
You haven't told us how you are fitting the model; are you using nls(), and if so with what initial values? The models don't make sense at x=0, due to the inclusion of the log(x) term. Ignoring that, you have 5 observations and 5 parameters in your second model. What is the reason you are including both b*log(x) and c*x terms in the model? regards albyn --- On Thu, Jan 12, 2006 at 07:11:12PM +0100, [EMAIL PROTECTED] wrote: Hi! I have a problem of curve fitting. I use the following data : - vector of predictor data : 0 0.4 0.8 1.2 1.6 - vector of response data : 0.81954 0.64592 0.51247 0.42831 0.35371 I perform parametric fits using custom equations when I use this equation : y = yo + K *(1/(1+exp(-(a+b*ln(x) the fitting result is OK but when I use this more general equation :y = yo + K *(1/(1+exp(-(a+b*log(x)+c*x , then I get an aberrant curve! I don't understand that... The second fitting should be at least as good as the first one because when taking c=0, both equations are identical! There is here a mathematical phenomenon that I don't understand!could someone help me Thanks a lot in advance! Nad?ge [[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 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] t-test for standard deviations
Sorry, Ted: Google on Brown-Forsythe and Levene's test and you will, indeed, find that rather robust and powerful t-tests can be used for testing homogeneity of spreads. In fact, on a variety of accounts, these tests are preferable to F-tests, which are notoriously non-robust (sensitive to non-normality) and which should long ago have been banned from statistics tects (IMHO). OTOH, whether one **should** test for homogeneity of spread instead of using statistical procedures robust to moderate heteroscedascity is another question. IMO, and I think on theoretical grounds, that is a better way to do things. Best yet is to use balanced designs in which most anything you do is less affected by any of these deviations from standard statistical assumptions. But that requires malice aforethought, rather than data dredging ... Cheers, Bert -- Bert Gunter Genentech Non-Clinical Statistics South San Francisco, CA The business of the statistician is to catalyze the scientific learning process. - George E. P. Box -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ted Harding Sent: Thursday, January 12, 2006 10:53 AM To: mirko sanpietrucci Cc: r-help@stat.math.ethz.ch Subject: Re: [R] t-test for standard deviations On 12-Jan-06 mirko sanpietrucci wrote: Dear R-users, I am new to the list and I would like to submit (probably) a stupid question: I found in a paper a reference to a t-test for the evaluationg the difference between the standard deviations of 2 samples. This test is performed in the paper but the methodology is not explained and any reference is reported. Does anyone know where I can find references to this test and if it is implemented in R? Thenks in advance for your help, Mirko If the paper says that a 1) t-test was used for evaluating the difference between the 2) standard deviations of 2 samples then I suspect that one or the other of these is a misprint. To compare standard deviations (more precisely, variances) you could use a (1)F-test. Or you would use a t-test to evaluate the difference between the (2)means of 2 samples. If it is really obscure what was done, perhaps an appropriate quotation from the paper would help to ascertain the problem. Best wishes, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 12-Jan-06 Time: 18:52:31 -- XFMail -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] bandwidth - Hmise.mixt - ks-package
Hello, I want to use Hmise.mixt for finding the optimal bandwidth. To be honest, I have only poor knowledge of the mathematical background. Using the help-file, I wrote: hopt-Hmise.mixt(c(0.0,1.5),c(1,1/9),c(0.75,0.25),100,0.4) but got the message: Error in ((i - 1) * d + 1):(i * d) : NA/NaN Argument Has anyone an idea, what's wrong? Thanks! Mala -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Basis of fisher.test
I want to ascertain the basis of the table ranking, i.e. the meaning of extreme, in Fisher's Exact Test as implemented in 'fisher.test', when applied to RxC tables which are larger than 2x2. One can summarise a strategy for the test as 1) For each table compatible with the margins of the observed table, compute the probability of this table conditional on the marginal totals. 2) Rank the possible tables in order of a measure of discrepancy between the table and the null hypothesis of no association. 3) Locate the observed table, and compute the sum of the probabilties, computed in (1), for this table and more extreme tables in the sense of the ranking in (2). The question is: what measure of discrepancy is used in 'fisher.test' corresponding to stage (2)? (There are in principle several possibilities, e.g. value of a Pearson chi-squared, large values being discrepant; the probability calculated in (2), small values being discrepant; ... ) ?fisher.test says only: In the one-sided 2 by 2 cases, p-values are obtained directly using the hypergeometric distribution. Otherwise, computations are based on a C version of the FORTRAN subroutine FEXACT which implements the network developed by Mehta and Patel (1986) and improved by Clarkson, Fan Joe (1993). The FORTRAN code can be obtained from URL: http://www.netlib.org/toms/643. I have had a look at this FORTRAN code, and cannot ascertain it from the code itself. However, there is a Comment to the effect: c PRE- Table p-value. (Output) c PRE is the probability of a more extreme table, where c 'extreme' is in a probabilistic sense. which suggests that the tables are ranked in order of their probabilities as computed in (2). Can anyone confirm definitively what goes on? With thanks, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 12-Jan-06 Time: 20:19:02 -- XFMail -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Multilevel models with mixed effects in R?
Yes, there are now multiple functions. One is the lmer() function in the
matrix package. Another is in the nlme package and is the lme()
function. Lmer is the newer version and the syntax has changed just
slightly. To see samples of the lmer function type the following at
your R command prompt
library(mlmRev)
vignette(MlmSoftRev)
This will open a pdf file with examples. You'll need to make sure to
obtain the mlmRev package from cran.
You can also see examples of student achievement analyses using these
functions in the following papers
http://cran.r-project.org/doc/Rnews/Rnews_2003-3.pdf
http://cran.r-project.org/doc/Rnews/Rnews_2005-1.pdf
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Charles Partridge
Sent: Thursday, January 12, 2006 2:54 PM
To: r-help@stat.math.ethz.ch
Subject: [R] Multilevel models with mixed effects in R?
Group,
I am new to R. In my work as a program evaluator, I am regularly asked
to estimate effect sizes of prevention/intervention and educational
programs on various student outcomes (e.g. academic achievement). In
many cases, I have access to data over three or more time periods (e.g.
growth in proficiency test scores).
I usually have multiple independent and dependent variables in each
model along with covariates. I have historically utilized latent growth
curve structural equation models, but would like to include random
effects in the model. Does R have the ability to run such analyses?
Regards,
Charles R. Partridge
Evaluation Specialist
Center for Learning Excellence
The John Glenn Institute for Public Service Public Policy
807 Kinnear Road, Room 214
Columbus, Ohio 43212-1421
Phone: 614.292.2419
FAX: 614.247.6447
Email: [EMAIL PROTECTED]
http://cle.osu.edu
CONFIDENTIALITY NOTICE: This message is intended only for\ t...{{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
Re: [R] Convert matrix to data.frame
When I try converting a matrix to a data frame, it works for me: x - matrix(1:6,ncol=2,dimnames=list(LETTERS[1:3],letters[24:25])) data.frame(x) x y A 1 4 B 2 5 C 3 6 str(data.frame(x)) `data.frame': 3 obs. of 2 variables: $ x: int 1 2 3 $ y: int 4 5 6 You can also use as.data.frame() to convert a matrix to a data.frame (but note that if colnames are missing form the matrix, as.data.frame() constructs different colnames than does data.frame(). You say it didn't work -- it's difficult to help with such a non-specific complaint. Can you explain exactly how it didn't work for you? (e.g., show the exact error message). -- Tony Plate Chia, Yen Lin wrote: Hi all, I wonder how could I convert a matrix A to a dataframe such that whenever I'm running a linear model such lme, I can use A$x1? I tried data.frame(A), it didn't work. Should I initialize A not as a matrix? Thanks. Yen Lin [[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 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Firths bias correction for log-linear models
Dear R-Help List, I'm trying to implement Firth's (1993) bias correction for log-linear models. Firth (1993) states that such a correction can be implemented by supplementing the data with a function of h_i, the diagonals from the hat matrix, but doesn't provide further details. I can see that for a saturated log-linear model, h_i=1 for all i, hence one just adds 1/2 to each count, which is equivalent to the Jeffrey's prior, but I'd also like to get bias corrected estimates for other log linear models. It appears that I need to iterate using GLM, with the weights option and h_i, which I can get from the function hatvalues. For logistic regression, this can be performed by splitting up each observation into response and nonresponse, and using weights as described in Heinze, G. and Schemper, M. (2002), but I'm unsure of how to implement the analogue for log-linear models. A procedure using IWLS is described by Firth (1992) in Dodge and Whittaker (1992), but this book isn't in the local library, and its $141+ on Amazon. I've tried looking at the code in the logistf and brlr libraries, but I haven't had any (successful) ideas. Can anyone help me in describing how to implement this in R? Thanks! Simon __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] CRAN versions of lme4/Matrix don't appear to work with R 2.1.1
Did you try upgrading to R 2.2.1? I just installed R 2.2.1 with the latest version of lme4 and Matrix, and they loaded fine for me. library(lme4) Loading required package: Matrix Loading required package: lattice sessionInfo() R version 2.2.1, 2005-12-20, i386-pc-mingw32 attached base packages: [1] methods stats graphics grDevices utils datasets [7] base other attached packages: lme4 latticeMatrix 0.98-1 0.12-11 0.99-6 Running Windows XP. (This may be obvious from 'i386-pc-mingw23', but I'm no sufficiently aware to know.) Doug Bates has worked very hard to create this, and I for one would not want to ask him to take the time to make it backward compatible for those who for whatever reason haven't yet upgraded to R 2.2.1, especially since for most of us it is fairly easy to upgrade. Spencer Graves White, Charles E WRAIR-Wash DC wrote: I am currently using R 2.1.1 under Windows and I do not seem to be able to load the current versions of lme4/Matrix. I have run 'update.packages.' I understand this is still experimental software but I would like access to a working version. Thanks. Chuck R Output: library(lme4) Loading required package: Matrix Error in lazyLoadDBfetch(key, datafile, compressed, envhook) : ReadItem: unknown type 241 In addition: Warning messages: 1: package 'lme4' was built under R version 2.3.0 2: package 'Matrix' was built under R version 2.3.0 Error: package 'Matrix' could not be loaded __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] I think not so hard question
I'm sorry to bother this list so much
But I haven't programmed in
A while and I'm struggling.
I have a vector in R of 1's and -1's
And I want to use a streak of size Y
To predict that the same value will
Be next.
So, suppose Y = 3. Then, if there is a streak of three
ones in a row, then I will predict that the next value is
a 1. But, if there is a streak of 3 -1's in a row,
then I will predict that a -1 is next. Otherwise,
I don't predict anything.
I am really new to R and kind of struggling
And I was wondering if someone could show
how to do this ?
In other words, given a vector of -1's
And 1's, I am unable ( I've been trying
For 2 days ) to create a new vector that
Has the predictions in it at the appropriate
places ? Thanks.
**
This email and any files transmitted with it are confidentia...{{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
Re: [R] t-test for standard deviations
On 12-Jan-06 Berton Gunter wrote: Sorry, Ted: Google on Brown-Forsythe and Levene's test and you will, indeed, find that rather robust and powerful t-tests can be used for testing homogeneity of spreads. In fact, on a variety of accounts, these tests are preferable to F-tests, which are notoriously non-robust (sensitive to non-normality) and which should long ago have been banned from statistics tects (IMHO). Not sure that I would consider either of these as a t-test as usually undestood. . Both are based on deriving a dispersion variable transform of the observations in each group (dquared or absolute deviation from the mean for Levene, absolute deivation from the mean for Brown-Forsythe), and performing an analysis of variance with the derived variable. Granted, in the case of two groups the ANOVA is equivalent to a squared t-test and one could indeed use a t-test instead of a 2-group ANOVA to get the directional information as well. But I would be surprised to find such a procedure referred to as a t-test as cited by Mirko. I think it would help if he told us a bit more about what the paper actually says. OTOH, whether one **should** test for homogeneity of spread instead of using statistical procedures robust to moderate heteroscedascity is another question. IMO, and I think on theoretical grounds, that is a better way to do things. Best yet is to use balanced designs in which most anything you do is less affected by any of these deviations from standard statistical assumptions. But that requires malice aforethought, rather than data dredging ... Your comments on the merits and advisability of these things are good -- but not forgetting that it is also a good idea to have enough understanding of what one is dealing with to be able to judge what might be best for the case in hand. However, I'm entirely with you when it comes to uncircumspect habitual use of standard procedures. Best wishes, Ted. Cheers, Bert -- Bert Gunter Genentech Non-Clinical Statistics South San Francisco, CA The business of the statistician is to catalyze the scientific learning process. - George E. P. Box -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ted Harding Sent: Thursday, January 12, 2006 10:53 AM To: mirko sanpietrucci Cc: r-help@stat.math.ethz.ch Subject: Re: [R] t-test for standard deviations On 12-Jan-06 mirko sanpietrucci wrote: Dear R-users, I am new to the list and I would like to submit (probably) a stupid question: I found in a paper a reference to a t-test for the evaluationg the difference between the standard deviations of 2 samples. This test is performed in the paper but the methodology is not explained and any reference is reported. Does anyone know where I can find references to this test and if it is implemented in R? Thenks in advance for your help, Mirko If the paper says that a 1) t-test was used for evaluating the difference between the 2) standard deviations of 2 samples then I suspect that one or the other of these is a misprint. To compare standard deviations (more precisely, variances) you could use a (1)F-test. Or you would use a t-test to evaluate the difference between the (2)means of 2 samples. If it is really obscure what was done, perhaps an appropriate quotation from the paper would help to ascertain the problem. Best wishes, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 12-Jan-06 Time: 18:52:31 -- XFMail -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 12-Jan-06 Time: 21:09:19 -- XFMail -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] bug with mai , pdf, and heatmap ?
Hi all, When using heatmap() with a pdf driver, and specifying parameters for mai in heatmap, I get a printout of the mai parameters at the top of the pdf...see attachment. This is on win2k pro with R2.2.1 Thanks, Ken __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Convert matrix to data.frame
Hi all, I wonder how could I convert a matrix A to a dataframe such that whenever I'm running a linear model such lme, I can use A$x1? I tried data.frame(A), it didn't work. Should I initialize A not as a matrix? Thanks. Yen Lin [[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
[R] February course *** R/Splus Fundamentals and Programming Techniques
XLSolutions Corporation ([1]www.xlsolutions-corp.com) is proud to
announce 2-day R/S-plus Fundamentals and Programming
Techniques in San Francisco: [2]www.xlsolutions-corp.com/Rfund.htm
San Francisco, February 16-17
Seattle,February 20-21
Boston,February 23-24
New York, February 27-28
Reserve your seat now at the early bird rates! Payment due AFTER
the class
Course Description:
This two-day beginner to intermediate R/S-plus course focuses on a
broad spectrum of topics, from reading raw data to a comparison of R
and S. We will learn the essentials of data manipulation, graphical
visualization and R/S-plus programming. We will explore statistical
data analysis tools,including graphics with data sets. How to enhance
your plots, build your own packages (librairies) and connect via
ODBC,etc.
We will perform some statistical modeling and fit linear regression
models. Participants are encouraged to bring data for interactive
sessions
With the following outline:
- An Overview of R and S
- Data Manipulation and Graphics
- Using Lattice Graphics
- A Comparison of R and S-Plus
- How can R Complement SAS?
- Writing Functions
- Avoiding Loops
- Vectorization
- Statistical Modeling
- Project Management
- Techniques for Effective use of R and S
- Enhancing Plots
- Using High-level Plotting Functions
- Building and Distributing Packages (libraries)
- Connecting; ODBC, Rweb, Orca via sockets and via Rjava
Email us for group discounts.
Email Sue Turner: [EMAIL PROTECTED]
Phone: 206-686-1578
Visit us: [4]www.xlsolutions-corp.com/training.htm
Please let us know if you and your colleagues are interested in this
classto take advantage of group discount. Register now to secure your
seat!
Interested in R/Splus Advanced course? email us.
Cheers,
Elvis Miller, PhD
Manager Training.
XLSolutions Corporation
206 686 1578
[5]www.xlsolutions-corp.com
[EMAIL PROTECTED]
References
1. http://www.xlsolutions-corp.com/
2. http://www.xlsolutions-corp.com/Rfund.htm
3. http://email.secureserver.net/view.php?folder=INBOXuid=2791#Compose
4. http://www.xlsolutions-corp.com/training.htm
5. http://www.xlsolutions-corp.com/
6. http://email.secureserver.net/view.php?folder=INBOXuid=2791#Compose
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Basis of fisher.test
(Ted Harding) [EMAIL PROTECTED] writes: I want to ascertain the basis of the table ranking, i.e. the meaning of extreme, in Fisher's Exact Test as implemented in 'fisher.test', when applied to RxC tables which are larger than 2x2. One can summarise a strategy for the test as 1) For each table compatible with the margins of the observed table, compute the probability of this table conditional on the marginal totals. 2) Rank the possible tables in order of a measure of discrepancy between the table and the null hypothesis of no association. 3) Locate the observed table, and compute the sum of the probabilties, computed in (1), for this table and more extreme tables in the sense of the ranking in (2). The question is: what measure of discrepancy is used in 'fisher.test' corresponding to stage (2)? (There are in principle several possibilities, e.g. value of a Pearson chi-squared, large values being discrepant; the probability calculated in (2), small values being discrepant; ... ) ?fisher.test says only: In the one-sided 2 by 2 cases, p-values are obtained directly using the hypergeometric distribution. Otherwise, computations are based on a C version of the FORTRAN subroutine FEXACT which implements the network developed by Mehta and Patel (1986) and improved by Clarkson, Fan Joe (1993). The FORTRAN code can be obtained from URL: http://www.netlib.org/toms/643. I have had a look at this FORTRAN code, and cannot ascertain it from the code itself. However, there is a Comment to the effect: c PRE- Table p-value. (Output) c PRE is the probability of a more extreme table, where c 'extreme' is in a probabilistic sense. which suggests that the tables are ranked in order of their probabilities as computed in (2). Can anyone confirm definitively what goes on? To my knowledge, it is the table probability, according to the hypergeometric distribution, i.e. the probability of the table given the marginals, which can be translated to sampling a+b balls without replacement from a box with a+c white and b+d black balls. Playing around with dhyper should be instructive. (You're right that the two-sided p values are obtained by summing all smaller or equal table probabilities. This is the traditional way, but there are alternatives, e.g. tail balancing.) -- 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
Re: [R] bug with mai , pdf, and heatmap ?
Ken Termiso wrote: Hi all, When using heatmap() with a pdf driver, and specifying parameters for mai in heatmap, I get a printout of the mai parameters at the top of the pdf...see attachment. This is on win2k pro with R2.2.1 Thanks, Ken Not a bug since ?heatmap has a main argument. Thus, heatmap(..., mai = c(1,2,3,4)) is actually interpretted as heatmap(..., main = c(1,2,3,4)) due to R's partial matching of the argument list. You should try: pdf() par(mai = c(1,2,3,4)) heatmap(...) dev.off() Hopefully I've assessed this correctly. HTH, --sundar P.S. See the posting guide regarding attachments to the list. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] CRAN versions of lme4/Matrix don't appear to work with R 2.1.1
I blew away my existing R directory structure, reinstalled R 2.2.1 from scratch, downloaded lme4 and associated packages from scratch, tried to load lme4 and got the same error message. Again, I am using the Windows version of R on XP. Chuck R : Copyright 2005, The R Foundation for Statistical Computing Version 2.2.1 (2005-12-20 r36812) ISBN 3-900051-07-0 library(lme4) Loading required package: Matrix Error in lazyLoadDBfetch(key, datafile, compressed, envhook) : ReadItem: unknown type 241 In addition: Warning messages: 1: package 'lme4' was built under R version 2.3.0 2: package 'Matrix' was built under R version 2.3.0 Error: package 'Matrix' could not be loaded __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] CRAN versions of lme4/Matrix don't appear to work with R 2.1.1
White, Charles E WRAIR-Wash DC [EMAIL PROTECTED] writes: I blew away my existing R directory structure, reinstalled R 2.2.1 from scratch, downloaded lme4 and associated packages from scratch, tried to load lme4 and got the same error message. Again, I am using the Windows version of R on XP. Which CRAN mirror? The Statlib one has been acting up lately. -- 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
Re: [R] Firths bias correction for log-linear models
On 12 Jan 2006, at 20:54, [EMAIL PROTECTED] wrote: Dear R-Help List, I'm trying to implement Firth's (1993) bias correction for log-linear models. Firth (1993) states that such a correction can be implemented by supplementing the data with a function of h_i, the diagonals from the hat matrix, but doesn't provide further details. I can see that for a saturated log-linear model, h_i=1 for all i, hence one just adds 1/2 to each count, which is equivalent to the Jeffrey's prior, but I'd also like to get bias corrected estimates for other log linear models. It appears that I need to iterate using GLM, with the weights option and h_i, which I can get from the function hatvalues. For logistic regression, this can be performed by splitting up each observation into response and nonresponse, and using weights as described in Heinze, G. and Schemper, M. (2002), but I'm unsure of how to implement the analogue for log-linear models. A procedure using IWLS is described by Firth (1992) in Dodge and Whittaker (1992), but this book isn't in the local library, and its $141+ on Amazon. I've tried looking at the code in the logistf and brlr libraries, but I haven't had any (successful) ideas. Can anyone help me in describing how to implement this in R? I don't recommend the adjusted IWLS approach in practice, because that algorithm is only first-order convergent. It is mainly of theoretical interest. The brlr function (in the brlr package) provides a template for a more direct approach in practice. The central operation there is an application of optim(), with objective function - (l + (0.5 * log(detinfo))) in which l is the log likelihood and detinfo is the determinant of the Fisher information matrix. In the case of a Poisson log-linear model, the Fisher information is, using standard GLM-type notation, t(X) %*% diag(mu) %*% X. It is straightforward to differentiate this penalized log-likelihood function, so (as in brlr) derivatives can be supplied for use with a second-order convergent optim() algorithm such as BFGS (see ?optim for a reference on the algorithm). I hope that helps. Please feel free to contact me off the list if anything is unclear. Kind regards, David -- Professor David Firth http://www.warwick.ac.uk/go/dfirth __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Curve fitting
ndurand at fr.abx.fr writes: Hi! I have a problem of curve fitting. I perform parametric fits using custom equations when I use this equation : y = yo + K *(1/(1+exp(-(a+b*ln(x) the fitting result is OK but when I use this more general equation :y = yo + K *(1/(1+exp(-(a+b*log(x)+c*x , then I get an aberrant curve! I don't understand that... The second fitting should be at least as good as the first one because when taking c=0, both equations are identical! Can you specify *exactly* what R code you're using? Are you using nls()? You're trying to fit a five-parameter model to five data points, which is likely to be difficult if not impossible to do statistically. Furthermore, your data points don't have very much information in them about all the parameters you're trying to estimate -- they are steadily decreasing, with very mild curvature. Finally, if these data happen to be points that you have generated as theoretical values, without adding noise, nls will give you problems (see ?nls). If you give us more detail about what you're trying to do we might be able to help (or possibly tell you that it really can't work ...) 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
[R] edit.data.frame
Dear list, Sometimes I have huge data.frames and the small spreadsheetlike edit.data.frame is quite handy to get an overview of the data. However, when I close the editor all data are rolled over the console window, which takes time and clutters the window. Is there a way to avoid this? Fredrik __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Data with no separator
I have data in which each row consists of a long string of number, letters, symbols, and blank spaces. I would like to simply scan in strings of length 426, but R takes the spaces that occur in the data as separators. Is there any way around this? Thanks, Jeff Steedle -- Jeffrey T. Steedle ([EMAIL PROTECTED]) Psychological Studies in Education Stanford University School of Education __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] CRAN versions of lme4/Matrix don't appear to work with R 2.1.1
It worked fine for me using http://cran.fhcrc.org/ Fred Hutchinson Cancer Research Center, Seattle, WA spencer graves Peter Dalgaard wrote: White, Charles E WRAIR-Wash DC [EMAIL PROTECTED] writes: I blew away my existing R directory structure, reinstalled R 2.2.1 from scratch, downloaded lme4 and associated packages from scratch, tried to load lme4 and got the same error message. Again, I am using the Windows version of R on XP. Which CRAN mirror? The Statlib one has been acting up lately. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Data with no separator
On 12-Jan-06 Jeffrey T. Steedle wrote: I have data in which each row consists of a long string of number, letters, symbols, and blank spaces. I would like to simply scan in strings of length 426, but R takes the spaces that occur in the data as separators. Is there any way around this? Thanks, Jeff Steedle You could use readLines(), perhaps? Data-readLines(datafile) should give you a vector Data of which each element is a character string which is one line read from your datafile. Best wishes, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 12-Jan-06 Time: 22:46:39 -- XFMail -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] CRAN versions of lme4/Matrix don't appear to work with R 2.1.1
I made a typographical error, since I am running R 2.2.1. I wouldn't dream of asking Professor Bates to make things backwards compatible to an old version. Since the packages are loading fine for you, I will assume there was a download error with the packages I have, uninstall them, and try again. Chuck -Original Message- From: Spencer Graves [mailto:[EMAIL PROTECTED] Sent: Thursday, January 12, 2006 4:08 PM To: White, Charles E WRAIR-Wash DC Cc: r-help@stat.math.ethz.ch; Douglas Bates Subject: Re: [R] CRAN versions of lme4/Matrix don't appear to work with R 2.1.1 Did you try upgrading to R 2.2.1? I just installed R 2.2.1 with the latest version of lme4 and Matrix, and they loaded fine for me. Doug Bates has worked very hard to create this, and I for one would not want to ask him to take the time to make it backward compatible for those who for whatever reason haven't yet upgraded to R 2.2.1, especially since for most of us it is fairly easy to upgrade. Spencer Graves __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] CRAN versions of lme4/Matrix don't appear to work with R 2.1.1
1) It was the Statlib mirror from which I was downloading. If I don't get any more interesting messages before I return to work in the morning, I'll try installing off of a different mirror. 2) On general principles, I just updated the lme4 related packages on the SuSE 10 machine that I run at home (NC mirror). That worked fine. Chuck -Original Message- From: [EMAIL PROTECTED] on behalf of Peter Dalgaard Sent: Thu 1/12/2006 4:56 PM Which CRAN mirror? The Statlib one has been acting up lately. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] CRAN versions of lme4/Matrix don't appear to work with R 2.1.1
Hi, all, This is interesting, since the problem I posted on a week ago was resolved by downloading (the current versions of) R and ctv from the Austria site; the UCLA site, from which I had downloaded before, had an old version of R labeled as the current version, and it was not compatible with the version of ctv that they had. Moral of story: It's not just the Statlib site that is (or at least can be) unreliable. Is it common knowledge that there are other mirrors which do not provide a very good reflection? Regards, Mark A. Dr. Mark C. Andersen Associate Professor Department of Fishery and Wildlife Sciences New Mexico State University Las Cruces NM 88003-0003 phone: 505-646-8034 fax: 505-646-1281 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of White, Charles E WRAIR-Wash DC Sent: Thursday, January 12, 2006 3:43 PM To: Peter Dalgaard Cc: Douglas Bates; r-help@stat.math.ethz.ch; Spencer Graves Subject: Re: [R] CRAN versions of lme4/Matrix don't appear to work with R 2.1.1 1) It was the Statlib mirror from which I was downloading. If I don't get any more interesting messages before I return to work in the morning, I'll try installing off of a different mirror. 2) On general principles, I just updated the lme4 related packages on the SuSE 10 machine that I run at home (NC mirror). That worked fine. Chuck -Original Message- From: [EMAIL PROTECTED] on behalf of Peter Dalgaard Sent: Thu 1/12/2006 4:56 PM Which CRAN mirror? The Statlib one has been acting up lately. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] jointprior in deal package
In case you haven't already solved this problem (I have seen no replies to this post), I will offer a suggestion. First, I've never used the deal package. I installed it and tried the example provided with the documentation for jointprior. It seemed to return something sensible -- certainly NOT an error message. Have you considered making a local copy of jointprior, then invoking 'debug(jointprior)', then walking trough the function line by line (as described in teh 'debug' documentation)? If you do this, you will find exactly the place the command that generates the error message. The 'debug' procedure also allows you to look at any object in the local environment created by jointprior. By doing this, you might get a better idea of the problem. Another alternative would be to consider the differences between the example provided with the documentation and your tor.nw. You may be able to identify the problem from that. If that failed, I would then try to modify tor.nw to produce the simplest possible example I could think of that would still produce the error message. In the course of doing that, you may be able to resolve the issue. If not, if you send your simplest possible example to this list, someone else might be able to help you. A reproducible example is nearly always easier to diagnose than a relatively vague description like you provided. To get an answer to the question that you asked, (a) your question must reach someone who has used the deal package and knows that specific error message and (b) that person must have the time and interest to respond. The probability of that happening may be quite low. By contrast, if you submit a toy example that doesn't quite work, anyone interested in playing a few minutes with the deal package can study your example and possibly take it to the next step. In general, I believe that providing a simple reproducible example probably increases by a couple of orders of magnitude the odds of receiving a useful answer quickly. hope this helps. spencer graves PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Tim Smits wrote: Dear all, I recently started using the deal package for learning Bayesian networks. When using the jointprior function on a particular dataset, I get the following message: tor.prior-jointprior(tor.nw) Error in array(1, Dim) : 'dim' specifies too large an array What is the problem? How can I resolve it? Thanks, Tim 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 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R - Wikis and R-core
[EMAIL PROTECTED] wrote: Hello Martin and others, I am happy with this decision. I'll look a little bit at this next week. Best, Philippe Grosjean We've had a small review time within R-core on this topic, amd would like to state the following: -- The R-core team welcomes proposals to develop an R-wiki. - We would consider linking a very small number of Wikis (ideally one) from www.r-project.org and offering an address in the r-project.org domain (such as 'wiki.r-project.org'). - The core team has no support time to offer, and would be looking for a medium-term commitment from a maintainer team for the Wiki(s). - Suggestions for the R documentation would best be filtered through the Wiki maintainers, who could e.g. supply suggested patches during the alpha phase of an R release. -- Our main concerns have been about ensuring the quality of such extra documentation projects, hence the 2nd point above. Several of our more general, not mainly R, experiences have been of outdated web pages which are continued to be used as reference when their advice has long been superseded. I think it's very important to try ensuring that this won't happen with an R Wiki. [Tried to send the following a few days ago, but had a problem with my connection:] What about adding a best before date on Wiki pages and let moderators extend such dates (by a simple click). If the date for a page is not updated, there will be a warning on that page telling the reader that the content might not be fully valid. MediaWiki is a good solution because there you can write equations in LaTeX, which are generated as Math-ML(?) and/or bitmap images. This feature might be in other wiki system too, I don't know. That's my $0.02 Henrik Martin Maechler, ETH Zurich PhGr == Philippe Grosjean [EMAIL PROTECTED] on Sun, 8 Jan 2006 17:00:44 +0100 (CET) writes: PhGr Hello all, Sorry for not taking part of this PhGr discussion earlier, and for not answering Detlef PhGr Steuer, Martin Maechler, and others that asked more PhGr direct questions to me. I am away from my office and PhGr my computer until the 16th of January. PhGr Just quick and partial answers: 1) I did not know PhGr about Hamburg RWiki. But I would be happy to merge PhGr both in one or the other way, as Detlef suggests it. PhGr 2) I choose DokuWiki as the best engine after a PhGr careful comparison of various Wiki engines. It is the PhGr best one, as far as I know, for the purpose of PhGr writting software documentation and similar PhGr pages. There is an extensive and clearly presented PhGr comparison of many Wikki engines at: PhGr http://www.wikimatrix.org/. PhGr 3) I started to change DokuWiki (addition of various PhGr plugins, addition of R code syntax coloring with PhGr GESHI, etc...). So, it goes well beyond all current PhGr Wiki engines regarding its suitability to present R PhGr stuff. PhGr 4) The reasons I did this is because I think the Wiki PhGr format could be of a wider use. I plan to change a PhGr little bit the DokuWiki syntax, so that it works with PhGr plain .R code files (Wiki part is simply embedded in PhGr commented lines, and the rest is recognized and PhGr formatted as R code by the Wiki engine). That way, the PhGr same Wiki document can either rendered by the Wiki PhGr engine for a nice presentation, or sourced in R PhGr indifferently. PhGr 5) My last idea is to add a Rpad engine to the Wiki, PhGr so that one could play with R code presented in the PhGr Wiki pages and see the effects of changes directly in PhGr the Wiki. PhGr 6) Regarding the content of the Wiki, it should be PhGr nice to propose to the authors of various existing PhGr document to put them in a Wiki form. Something like PhGr Statistics with R PhGr (http://zoonek2.free.fr/UNIX/48_R/all.html) is written PhGr in a way that stimulates additions to pages in PhGr perpetual construction, if it was presented in a Wiki PhGr form. It is licensed as Creative Commons PhGr Attribution-NonCommercial-ShareAlike 2.5 license, that PhGr is, exactly the same one as DokuWiki that I choose for PhGr R Wiki. Of course, I plan to ask its author to do so PhGr before putting its hundreds of very interesting pages PhGr on the Wiki... I think it is vital to have already PhGr something in the Wiki, in order to attract enough PhGr readers, and then enough contributors! PhGr 7) Regarding spamming and vandalism, DokuWiki allows PhGr to manage rights and users, even individually for PhGr pages. I think it would be fine to lock pages that PhGr reach a certain maturity (read-only / editable by PhGr selected users only) , with link to a
Re: [R] CRAN versions of lme4/Matrix don't appear to work with R 2.1.1
I don't know, but I had a similar problem a few weeks ago with the one or both of the California sites (I don't remember which now). The problems disappeared after I switched to http://cran.fhcrc.org/; (Fred Hutchinson Cancer Research Center, Seattle, WA). spencer graves Mark Andersen wrote: Hi, all, This is interesting, since the problem I posted on a week ago was resolved by downloading (the current versions of) R and ctv from the Austria site; the UCLA site, from which I had downloaded before, had an old version of R labeled as the current version, and it was not compatible with the version of ctv that they had. Moral of story: It's not just the Statlib site that is (or at least can be) unreliable. Is it common knowledge that there are other mirrors which do not provide a very good reflection? Regards, Mark A. Dr. Mark C. Andersen Associate Professor Department of Fishery and Wildlife Sciences New Mexico State University Las Cruces NM 88003-0003 phone: 505-646-8034 fax: 505-646-1281 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of White, Charles E WRAIR-Wash DC Sent: Thursday, January 12, 2006 3:43 PM To: Peter Dalgaard Cc: Douglas Bates; r-help@stat.math.ethz.ch; Spencer Graves Subject: Re: [R] CRAN versions of lme4/Matrix don't appear to work with R 2.1.1 1) It was the Statlib mirror from which I was downloading. If I don't get any more interesting messages before I return to work in the morning, I'll try installing off of a different mirror. 2) On general principles, I just updated the lme4 related packages on the SuSE 10 machine that I run at home (NC mirror). That worked fine. Chuck -Original Message- From: [EMAIL PROTECTED] on behalf of Peter Dalgaard Sent: Thu 1/12/2006 4:56 PM Which CRAN mirror? The Statlib one has been acting up lately. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] hypothesis testing for rank-deficient linear models
[EMAIL PROTECTED] wrote:
Take the following example:
a - rnorm(100)
b - trunc(3*runif(100))
g - factor(trunc(4*runif(100)),labels=c('A','B','C','D'))
y - rnorm(100) + a + (b+1) * (unclass(g)+2)
...
Here's a cleaned-up function to compute estimable within-group effects
for rank deficient models. For the above data, it could be invoked
as:
m - lm(y~a+b*g, subset=(b==0 | g!='B'))
subgroup.effects(m, 'b', g=c('A','B','C','D'))
g Estimate StdError t.value p.value
1 A 2.779167 0.4190213 6.63252 4.722978e-09
2 B NANA NA NA
3 C 4.572431 0.3074402 14.87258 6.226445e-24
4 D 5.920809 0.3502251 16.90572 3.995266e-27
Again, I'm not sure whether this is a good approach, or whether there
is an easier way using existing R functions. One problem is figuring
exactly which terms are not estimable from the available data. My
hack using alias() is not satisfactory and I've already run into cases
where it fails. But I'm having trouble coming up with a more general,
correct test?
-- David Hinds
subgroup.effects - function(model, term, ...)
{
my.coef - function(n)
{
contr - lapply(names(args), function(i)
contr.treatment(args[[i]], unclass(gr[n,i])))
names(contr) - names(args)
u - update(model, formula=model$formula,
data=model$data, contrasts=contr)
uc - coef(summary(u))[term,]
if (any(is.na(coef(u)))
any(!is.na(alias(u)$Complete)))
uc[1:4] - NA
uc
}
args - list(...)
gr - expand.grid(...)
d - data.frame(gr, t(sapply(1:nrow(gr), my.coef)))
names(d) - c(names(gr),'Estimate','StdError','t.value','p.value')
d
}
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Basis of fisher.test
On Thu, 12 Jan 2006 [EMAIL PROTECTED] wrote: I want to ascertain the basis of the table ranking, i.e. the meaning of extreme, in Fisher's Exact Test as implemented in 'fisher.test', when applied to RxC tables which are larger than 2x2. One can summarise a strategy for the test as 1) For each table compatible with the margins of the observed table, compute the probability of this table conditional on the marginal totals. 2) Rank the possible tables in order of a measure of discrepancy between the table and the null hypothesis of no association. 3) Locate the observed table, and compute the sum of the probabilties, computed in (1), for this table and more extreme tables in the sense of the ranking in (2). The question is: what measure of discrepancy is used in 'fisher.test' corresponding to stage (2)? (There are in principle several possibilities, e.g. value of a Pearson chi-squared, large values being discrepant; the probability calculated in (2), small values being discrepant; ... ) ?fisher.test says only: [That following is not a quote from a current version of R.] In the one-sided 2 by 2 cases, p-values are obtained directly using the hypergeometric distribution. Otherwise, computations are based on a C version of the FORTRAN subroutine FEXACT which implements the network developed by Mehta and Patel (1986) and improved by Clarkson, Fan Joe (1993). The FORTRAN code can be obtained from URL: http://www.netlib.org/toms/643. No, it *also* says Two-sided tests are based on the probabilities of the tables, and take as 'more extreme' all tables with probabilities less than or equal to that of the observed table, the p-value being the sum of such probabilities. which answers the question (there are only two-sided tests for such tables). Now, what does the posting guide say about stating the R version and updating before posting? -- 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
Re: [R] edit.data.frame
On Thu, 12 Jan 2006, Fredrik Lundgren wrote: Sometimes I have huge data.frames and the small spreadsheetlike edit.data.frame is quite handy to get an overview of the data. However, when I close the editor all data are rolled over the console window, which takes time and clutters the window. Is there a way to avoid this? If you mean printed to the R terminal/console, assign the result or use invisible(edit(object)). -- 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
Re: [R] edit.data.frame
I think fix(data.frame.name) is the best way . 2006/1/13, Prof Brian Ripley [EMAIL PROTECTED]: On Thu, 12 Jan 2006, Fredrik Lundgren wrote: Sometimes I have huge data.frames and the small spreadsheetlike edit.data.frame is quite handy to get an overview of the data. However, when I close the editor all data are rolled over the console window, which takes time and clutters the window. Is there a way to avoid this? If you mean printed to the R terminal/console, assign the result or use invisible(edit(object)). -- 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 -- 黄荣贵 Deparment of Sociology Fudan University __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Taking code from packages
Hello! I am currently in the process of creating (my first) package, which (when ready) I intend to publish to CRAN. In the process of creating this package I have taken some code form existing packages. I have actually copied parts of functions in to new functions. This code is usually something very basic such as Rand index. What is the proper procedure for this? Since most of R (and also the packages I have taken code form) is published under GPL, I think this should be OK. However I do not know if: 1. I should still ask authors of the packages for permission or at least notify them. 2. Ad references to the functions (and packages) from which I had taken the code or only to the references they use. What about regarding code that was sent to the list, usually as a response to one of my problems. I assume that in this case it is best to consult the author? Any comments and opinions are very welcomed! Best regards, Ales Ziberna __ 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
