Re: [R] effect sizes for Wilcoxon tests
On Tue, 15 Nov 2005, Thomas Lumley wrote: On Tue, 15 Nov 2005, Claus Atzenbeck wrote: On Tue, 15 Nov 2005, Peter Dalgaard wrote: However, for Wilcoxon tests, the formula for effect sizes is: r = Z / sqrt(N) I wonder how I can calculate the Z-score in R for a Wilcoxon test. Does anyone know how to calculate effect sizes for Wilcoxon tests as SPSS can do? I wonder if it would be very difficult to do so with R. If the above formula is correct, it can't be hard. What I wonder is whether (and if so, how) it makes sense. ( the fact that SPSS does it is no guarantee...) According to my references, the formula is OK. I think you have misinterpreted the direction of Peter's scepticism. The question is whether an effect size defined this way means anything useful. For example, even if your data are Normally distributed with equal variance this definition will not agree with the definition based on the mean. This is only the start of the potential problems, especially if the distributions do not have the same shape. However, how do I get Z from a Wilcoxon test in R? wtest - wilcox.test(y~group,data=d, alternative=greater) qnorm(wtest$p.value) or library(coin) statistic(wilcox_test(y ~ group, data = d, ...), type = standardized) where the variance `estimator' takes care of tied observations. Best, Torsten -thomas __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] effect sizes for Wilcoxon tests
On Tue, 15 Nov 2005, Thomas Lumley wrote: I think you have misinterpreted the direction of Peter's scepticism. The question is whether an effect size defined this way means anything useful. For example, even if your data are Normally distributed with equal variance this definition will not agree with the definition based on the mean. This is only the start of the potential problems, especially if the distributions do not have the same shape. I see the point. The book that I use (it is a kind of practical guide) calculates the effect sizes of Mann-Whitney tests and compares it to Cohen's benchmarks (small effect: r=.10 -- medium effect: r=.30 -- large effect: r=.50). This was also my plan. However, as far as I understand you and Peter, the formula is not appropriate for comparison with Cohen's benchmarks? However, how do I get Z from a Wilcoxon test in R? wtest - wilcox.test(y~group,data=d, alternative=greater) qnorm(wtest$p.value) Thanks, also to Peter. Claus __ R-help@stat.math.ethz.ch 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] www.krankenversicherung.ch News - Krankenkassen - Information Newsletter
Hans-Peter == Hans-Peter [EMAIL PROTECTED] on Tue, 15 Nov 2005 16:10:35 +0100 writes: Hans-Peter Am 14.11.05 schrieb [EMAIL PROTECTED] [EMAIL PROTECTED]: [snip some spam] Hans-Peter Being located only some blocks away from these spammers, I just asked Hans-Peter them in person to remove the R list from their database. Vielen Dank, Hans-Peter. Es betrifft auch noch andere R-..@stat.math.ethz.ch Listen. Und dann könnte man doch gleich alle @...ethz.ch auch streichen... Aber eigentlich müssten die davon überzeugt werden können, nicht mehr zu spammen. Mit liebem Gruss, Martin Mächler (R- Mailing List Administrator) Martin [EMAIL PROTECTED] http://stat.ethz.ch/people/maechler Seminar fuer Statistik, ETH-Zentrum LEO C16Leonhardstr. 27 ETH (Federal Inst. Technology) 8092 Zurich SWITZERLAND phone: +41-44-632-3408 fax: ...-1228 __ R-help@stat.math.ethz.ch 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] spatial statistics on images, any packages?
Hi list, Is there a package that covers the evaluation of spatial statistics on images and not on point data? I've converted an image matrix to x, y coordinates and a measurement value but evaluation with the package spdep (not really designed for image data I suppose) is unworkable. Any suggestions? Regards, Koen __ R-help@stat.math.ethz.ch 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] matrix assembly
Hi I have a function f(k,l) which returns a matrix for integer k and l. I want to call f(.,.) for each combination of a supplied vector (as in expand.grid()) and then I want to assemble these matrices into one big one using k and l to index the position of the submatrices returned by f(k,l). Toy example follows. f - function(k,l){ matrix(k+l , k , l) } and I want to call f() with each combination of c(1,3) for the first argument and c(2,4,5) for the second and then assemble the resulting matrices. I can do it piecemeal: a - c(1,3) b - c(2,4,5) expand.grid(a,b) Var1 Var2 112 232 314 434 515 635 cbind(rbind(f(1,2),f(3,2)) , rbind(f(1,4),f(3,4)) , rbind(f(1,5),f (3,5))) [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [1,]335555666 6 6 [2,]557777888 8 8 [3,]557777888 8 8 [4,]557777888 8 8 [see how the calls to f(. , .) follow the rows of expand.grid(a,b)] How to do this in a nice vectorized manner? -- Robin Hankin Uncertainty Analyst National Oceanography Centre, Southampton European Way, Southampton SO14 3ZH, UK tel 023-8059-7743 __ R-help@stat.math.ethz.ch 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] effect sizes for Wilcoxon tests
Torsten Hothorn [EMAIL PROTECTED] writes: [snip] However, how do I get Z from a Wilcoxon test in R? wtest - wilcox.test(y~group,data=d, alternative=greater) qnorm(wtest$p.value) or library(coin) statistic(wilcox_test(y ~ group, data = d, ...), type = standardized) where the variance `estimator' takes care of tied observations. Doesn't it do that in the same way as inside wilcox.test(...,exact=FALSE)? Just wondering. -p -- 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] Histogram font
The hist() command produces this message on my machine... Error in title(main = main, sub = sub, xlab = xlab, ylab = ylab, ...) : X11 font at size 14 could not be loaded How can I either (a) determine what font is required, or (b) specify one of the fonts I have available? This problem is specific to hist(), plot() works fine. I am using R on SuSE Linux v9.3, from the KDE desktop. -- Christopher Willmot __ R-help@stat.math.ethz.ch 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] www.... News - Krankenkassen - ....
Sorry to everyone -- I'm deeply embarrassed that this private reply (to Hans-Peter) accidentally went to the whole R-help list. It's bad enough (and inavoidable as long as we keep the list open) that real spam occasionally goes through, but of all people, I should know better and be more careful Martin Maechler, ETH Zurich __ R-help@stat.math.ethz.ch 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] effect sizes for Wilcoxon tests
On Wed, 16 Nov 2005, Peter Dalgaard wrote: Torsten Hothorn [EMAIL PROTECTED] writes: [snip] However, how do I get Z from a Wilcoxon test in R? wtest - wilcox.test(y~group,data=d, alternative=greater) qnorm(wtest$p.value) or library(coin) statistic(wilcox_test(y ~ group, data = d, ...), type = standardized) where the variance `estimator' takes care of tied observations. Doesn't it do that in the same way as inside wilcox.test(...,exact=FALSE)? My understanding was that `wilcox.test' implements the unconditional version (with unconditional variance estimator and some `adjustment' for ties) and `wilcox_test' implements the conditional version of the test (of course both coincide when there are no ties). However, some quick experiments suggest that the standardized statistic is the same for both versions (with correct = FALSE) for tied observations. One needs to check if the expectation and variance formulae in `wilcox.test' are equivalent with the conditional versions used in `wilcox_test' (in contrast to my initial opinion). Best, Torsten Just wondering. -p -- O__ Peter Dalgaard ?ster 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] natural selection coefficient S
Hi everyone, before I embark in some coding, can I ask if there is any function in any of the packages on CRAN dealing with the natural selection coefficient S? At the moment I am more curious to see it has been implemented already in R, rather than what specific implementation. Regards, Federico Calboli -- Federico C. F. Calboli Department of Epidemiology and Public Health Imperial College, St Mary's Campus Norfolk Place, London W2 1PG Tel +44 (0)20 7594 1602 Fax (+44) 020 7594 3193 f.calboli [.a.t] imperial.ac.uk f.calboli [.a.t] gmail.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] Friedman test with replicated samples
Dear R-users, I would need to do a Friedman test with replicated samples, and Friedman.test(y...) currently works only for unreplicated designs. Is there a script or a function available? Thanks, Alessandro -- -- __ R-help@stat.math.ethz.ch 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] y-axis in histograms
Michael Graber wrote: Dear R- list, I have some data to present with histograms. Therefore I used hist(...). I have few values with almost 80% of the frequencies (totaly 800) and some other values with low frequencies ( totaly 5 -10 ) that I want to emphasize. Therefore I want to cut the y-axis on 100, but I don't know how to deal with this. If you mean that you would like to have an axis break from 10 to 100, you can try this: # be creative, make up some data fakedat-c(sample(1:6,50,TRUE),sample(7:9,1000,TRUE)) # get the histogram counts # note that I have to explicitly specify breaks fakehist-hist(fakedat,breaks=0:10,type=n) # here's a rough function gap.barplot-function(y,gap,x=NA,xlabels=NA,col=NULL,...) { if(missing(y) || missing(gap)) stop(Must pass at least y and gap) littleones-which(y=gap[1]) bigones-which(y=gap[2]) if(length(bigones)+length(littleones) != length(y)) warning(gap includes some values of y) gapsize-gap[2]-gap[1] if(is.na(x)) x-1:length(y) if(is.na(xlabels)) xlabels-as.character(x) xlim-range(x) ylim-c(min(y),max(y)-gapsize) plot(0,xlim=xlim,ylim=ylim,axes=FALSE,type=n,...) box() axis(1,at=x,labels=xlabels) ytics-pretty(y) littletics-which(yticsgap[1]) bigtics-which(ytics=gap[2]) axis(2,at=c(ytics[littletics],ytics[bigtics]-gapsize), labels=c(ytics[littletics],ytics[bigtics])) axis.break(2,gap[1]) halfwidth-min(diff(x))/2 plot.lim-par(usr) rect(x-halfwidth,plot.lim[3],x+halfwidth, c(y[littleones],y[bigones]-gapsize), col=col) abline(h=gap[1],col=white,lwd=5) } gap.barplot(fakehist$counts,gap=c(100,295)) Jim __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] spatial statistics on images, any packages?
On Wed, 16 Nov 2005, Koen Hufkens wrote: Hi list, Is there a package that covers the evaluation of spatial statistics on images and not on point data? I've converted an image matrix to x, y coordinates and a measurement value but evaluation with the package spdep (not really designed for image data I suppose) is unworkable. Any suggestions? It depends what you are trying to do, also please consider re-posting on the R-sig-geo list. Yes, constructing a neighbour list for a large image - millions of points and thus potentially very many neighbours - is not what spdep is dimensioned for. Advice will depend on your research problem. Regards, Koen -- 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
Re: [R] Histogram font
On Wed, 16 Nov 2005, Christopher Willmot wrote: The hist() command produces this message on my machine... Error in title(main = main, sub = sub, xlab = xlab, ylab = ylab, ...) : X11 font at size 14 could not be loaded How can I either (a) determine what font is required, or (b) specify one of the fonts I have available? This problem is specific to hist(), plot() works fine. I am using R on SuSE Linux v9.3, from the KDE desktop. Try searching the mailing list, this feels like: http://finzi.psych.upenn.edu/R/Rhelp02a/archive/20511.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
Re: [R] Replace missing values in spatial design using moving average
On Wed, 9 Nov 2005, Arnaud Dowkiw wrote: Dera R helpers, I have a (x,y,z) data file where x and y are spatial coordinates and z a variable. I have some missing values in the z column and I would like to replace them with an optimized estimation from the neighbour cells. I could not find any function in R to do that. Is anybody aware of such function ? Maybe someone knows how to implement the Papadakis method in R... Thanks for your help, I'm not sure about the Papadakis method, I don't think it is there, though could be implemented using standard tools. Ripley (1981) relates the method to SAR and CAR models, but predicting with them is far from easy (there are papers by Martin from the 1980's about this). Are geostatistical methods an option for you? That is, what is the support of your cells? Arnaud DOWKIW - - - - - - - - - - - - - - - - - - - - - - - Arnaud DOWKIW INRA Unité Amélioration Génétique et Physiologie Forestières 2163 avenue de la Pomme de Pin BP 20619 ARDON 45166 OLIVET CEDEX FRANCE Tel. + 33 2 38 41 78 00 Fax. + 33 2 38 41 48 09 - - - - - - - - - - - - - - - - - - - - - - - __ R-help@stat.math.ethz.ch 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
Re: [R] effect sizes for Wilcoxon tests
[EMAIL PROTECTED] writes: On Wed, 16 Nov 2005, Peter Dalgaard wrote: Torsten Hothorn [EMAIL PROTECTED] writes: [snip] However, how do I get Z from a Wilcoxon test in R? wtest - wilcox.test(y~group,data=d, alternative=greater) qnorm(wtest$p.value) or library(coin) statistic(wilcox_test(y ~ group, data = d, ...), type = standardized) where the variance `estimator' takes care of tied observations. Doesn't it do that in the same way as inside wilcox.test(...,exact=FALSE)? My understanding was that `wilcox.test' implements the unconditional version (with unconditional variance estimator and some `adjustment' for ties) and `wilcox_test' implements the conditional version of the test (of course both coincide when there are no ties). However, some quick experiments suggest that the standardized statistic is the same for both versions (with correct = FALSE) for tied observations. One needs to check if the expectation and variance formulae in `wilcox.test' are equivalent with the conditional versions used in `wilcox_test' (in contrast to my initial opinion). I think you'll find that they are the same. There isn't really an unconditional variance formula in the presence of ties - I don't think you can do that without knowing what the point masses are in the underlying distribution. The question is only whether the tie corrected statistic is an asymptotic approximation or an exact formula for the variance. I believe it is the latter. What you need to calculate is the expectation and variance of the (possibly tied) rank of a particular observation, given the sets of tied observations. In principle, also the covariance between two of them, but this is easily seen to be equal to -1/(N-1) times the variance since they are all equal and the rows/columns of the covariance sums to zero. The expectation is a no-brainer: tie-breaking preserves the sum of ranks so the average rank is left unchanged by ties. The fun bit is trying to come up with an elegant argument why the correction term for the variances, involving sum(NTIES.CI^3 - NTIES.CI) is exact. I think you can do it by saying that breaking a set of tied ranks randomly corresponds to adding a term which has a variance related to that of a random number between 1 and d, with probability d/N . Notice that sum((1:d)^2) - sum(1:d)^2 is (d^3-d)/3. After breaking the ties at random, you should end up with the untied situation, so you get the tied variance by subtracting the variance of the tie-breaking terms. Tying up the loose ends is left as an exercise Best, Torsten Just wondering. -p -- 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 -- 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] matrix assembly
a crude approach is the following: f - function(x){ k - x[1] l - x[2] matrix(k + l, k , l) } a - c(1, 3) b - c(2, 4, 5) combs - expand.grid(a, b) do.call(cbind, lapply(split(combs, combs$Var2), function(x) do.call(rbind, apply(x, 1, f 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: Robin Hankin [EMAIL PROTECTED] To: RHelp r-help@stat.math.ethz.ch Sent: Wednesday, November 16, 2005 11:10 AM Subject: [R] matrix assembly Hi I have a function f(k,l) which returns a matrix for integer k and l. I want to call f(.,.) for each combination of a supplied vector (as in expand.grid()) and then I want to assemble these matrices into one big one using k and l to index the position of the submatrices returned by f(k,l). Toy example follows. f - function(k,l){ matrix(k+l , k , l) } and I want to call f() with each combination of c(1,3) for the first argument and c(2,4,5) for the second and then assemble the resulting matrices. I can do it piecemeal: a - c(1,3) b - c(2,4,5) expand.grid(a,b) Var1 Var2 112 232 314 434 515 635 cbind(rbind(f(1,2),f(3,2)) , rbind(f(1,4),f(3,4)) , rbind(f(1,5),f (3,5))) [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [1,]335555666 6 6 [2,]557777888 8 8 [3,]557777888 8 8 [4,]557777888 8 8 [see how the calls to f(. , .) follow the rows of expand.grid(a,b)] How to do this in a nice vectorized manner? -- Robin Hankin Uncertainty Analyst National Oceanography Centre, Southampton European Way, Southampton SO14 3ZH, UK tel 023-8059-7743 __ R-help@stat.math.ethz.ch 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
[R] numericDeriv
I have to compute some standard errors using the delta method and so have to use the command numericDeriv to get the desired gradient. Befor using it on my complicated function, I've done a try with a simple exemple : x - 1:5 numericDeriv(quote(x^2),x) and i get : [1] 1 8 27 64 125 216 attr(,gradient) [,1] [,2] [,3] [,4] [,5] [,6] [1,] Inf00 NaN00 [2,]000 NaN00 [3,]0 Inf0 NaN00 [4,]000 NaN00 [5,]00 Inf NaN00 [6,]000 NaN00 I don't understand the result. I thought I will get : [1] 1 8 27 64 125 216 attr(,gradient) [,1] [1,] 1 [2,] 4 [3,] 6 [4,] 8 [5,] 10 [6,] 12 The derivative of x^2 is still 2x, isn't it ? Thanks for help __ R-help@stat.math.ethz.ch 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] Histogram font
On Wed, 16 Nov 2005, Christopher Willmot wrote: The hist() command produces this message on my machine... Error in title(main = main, sub = sub, xlab = xlab, ylab = ylab, ...) : X11 font at size 14 could not be loaded How can I either (a) determine what font is required, or (b) specify one of the fonts I have available? ?options (see option X11fonts). This problem is specific to hist(), plot() works fine. I am using R on SuSE Linux v9.3, from the KDE desktop. This usually means that 100dpi fonts are missing or not in the X11 font path. The code assumes that the standard 75dpi and where necessary 100dpi fonts are installed. The issue is title(), not hist() and not plot() as the error message shows. So calling plot(main=) would also fail. -- 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] matrix assembly
On 11/16/05, Robin Hankin [EMAIL PROTECTED] wrote: Hi I have a function f(k,l) which returns a matrix for integer k and l. I want to call f(.,.) for each combination of a supplied vector (as in expand.grid()) and then I want to assemble these matrices into one big one using k and l to index the position of the submatrices returned by f(k,l). Toy example follows. f - function(k,l){ matrix(k+l , k , l) } and I want to call f() with each combination of c(1,3) for the first argument and c(2,4,5) for the second and then assemble the resulting matrices. I can do it piecemeal: a - c(1,3) b - c(2,4,5) expand.grid(a,b) Var1 Var2 112 232 314 434 515 635 cbind(rbind(f(1,2),f(3,2)) , rbind(f(1,4),f(3,4)) , rbind(f(1,5),f (3,5))) [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [1,]335555666 6 6 [2,]557777888 8 8 [3,]557777888 8 8 [4,]557777888 8 8 [see how the calls to f(. , .) follow the rows of expand.grid(a,b)] How to do this in a nice vectorized manner? Replacing your example with the appropriate do.call and lapply calls gives this: do.call(cbind, lapply(b, function(x) do.call(rbind, lapply(a, f, l = x or one can do the cbind in the inner loop and the rbind in the outer giving the same result: do.call(rbind, lapply(a, function(x) do.call(cbind, lapply(b, f, k = x __ R-help@stat.math.ethz.ch 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] www.... News - Krankenkassen - ....
Hello Martin, 2005/11/16, Martin Maechler [EMAIL PROTECTED]: Sorry to everyone -- I'm deeply embarrassed that this private reply (to Hans-Peter) accidentally went to the whole R-help list. It's bad enough (and inavoidable as long as we keep the list open) that real spam occasionally goes through, but of all people, I should know better and be more careful In the end there is only a very small segment of spam getting the spam filters, not to say: Great job, no reason to worry. After all, finally everone has the possiblity to also kill spam in its only mail client aterwards! On this occasion, thanks for setting up and running this good and helpful list. sincerely Thomas __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] numericDeriv
On Wed, 16 Nov 2005, Florent Bresson wrote: I have to compute some standard errors using the delta method and so have to use the command numericDeriv to get the desired gradient. Befor using it on my complicated function, I've done a try with a simple exemple : x - 1:5 numericDeriv(quote(x^2),x) and i get : [1] 1 8 27 64 125 216 attr(,gradient) [,1] [,2] [,3] [,4] [,5] [,6] [1,] Inf00 NaN00 [2,]000 NaN00 [3,]0 Inf0 NaN00 [4,]000 NaN00 [5,]00 Inf NaN00 [6,]000 NaN00 I don't understand the result. I thought I will get : [1] 1 8 27 64 125 216 attr(,gradient) [,1] [1,] 1 [2,] 4 [3,] 6 [4,] 8 [5,] 10 [6,] 12 The derivative of x^2 is still 2x, isn't it ? and (1:5)^2 is still [1] 1 4 9 16 25 ! Try x - as.numeric(1:5) numericDeriv(quote(x^2),x) since the author of numericDeriv has forgotten some coercions. -- 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] numericDeriv
On 16-Nov-05 Florent Bresson wrote: I have to compute some standard errors using the delta method and so have to use the command numericDeriv to get the desired gradient. Befor using it on my complicated function, I've done a try with a simple exemple : x - 1:5 numericDeriv(quote(x^2),x) and i get : [1] 1 8 27 64 125 216 attr(,gradient) [,1] [,2] [,3] [,4] [,5] [,6] [1,] Inf00 NaN00 [2,]000 NaN00 [3,]0 Inf0 NaN00 [4,]000 NaN00 [5,]00 Inf NaN00 [6,]000 NaN00 I don't understand the result. I thought I will get : [1] 1 8 27 64 125 216 attr(,gradient) [,1] [1,] 1 [2,] 4 [3,] 6 [4,] 8 [5,] 10 [6,] 12 The derivative of x^2 is still 2x, isn't it ? The trap you've fallen into is that x - 1:5 makes x of integer type, and (believe it or not) you cannot differentiate when the support of a function is the integers. Wrong topology (though I'm not sure that this is quite how R thinks about it). So give x a bit of elbow-room (numeric type has continous -- well, nearly -- topology): x - as.numeric(1:5) numericDeriv(quote(x^2),x) [1] 1 4 9 16 25 attr(,gradient) [,1] [,2] [,3] [,4] [,5] [1,]20000 [2,]04000 [3,]00600 [4,]00080 [5,]0000 10 Cheers, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 16-Nov-05 Time: 13:11:49 -- 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] Difficulties with for() {while(){}}
Is this possible: function(){ initialize - function( ){ initialize } for(i in test){ ... if( j = test2 ) {i - 1; initialize()} ... } Best, Matthias Hi, I have the follow function: function() { ## Init of function ... for(i in test) { ... while(j = test2) { ... } } } The problem is that sometimes, naturally, the while is not possible to be resolved, and so the program abort. In this case I need that program return to the init of function and run again. How I can make this? Abort the while, abort the for and run the function again? Thanks Ronaldo -- Os homens ficam terrivelmente chatos quando sao bons maridos, e abominavelmente convencidos quando nao sao. -- Oscar Wilde -- | // | \\ [***] | ( õ õ ) [Ronaldo Reis Júnior] | V [UFV/DBA-Entomologia] |/ \ [36570-000 Viçosa - MG ] | /(.''`.)\ [Fone: 31-3899-4007 ] | /(: :' :)\ [EMAIL PROTECTED]] |/ (`. `'` ) \[ICQ#: 5692561 | LinuxUser#: 205366 ] |( `- ) [***] | _/ \_Powered by GNU/Debian Woody/Sarge __ R-help@stat.math.ethz.ch 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] Error in integrate
Hi! I am a beginner of R. I am trying to calculate integrate and draw a graph of the output, but just kept on getting error messages. I list my program and error message below. Please help. Many Thanks! === + a--11 b-0.1 c-0.012 x-0:110 t-0:15 integrand-function(x) {exp(-exp(a-c*t)*(exp(b*x)-exp(c*x))/(b-c))} cal-integrate(integrand,0,Inf) Error in integrate(integrand, 0, Inf) : evaluation of function gave a result of wrong length In addition: Warning message: longer object length is not a multiple of shorter object length in: -exp(a - c * t) * (exp(b * x) - exp(c * x)) plot(cal~t, xlab=Time, ylab=Pop at t) Error in model.frame(formula, rownames, variables, varnames, extras, extranames, : invalid variable type === - Vivien Chen - [[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] Combine related plots
On 11/16/05 8:48 AM, Vivien W. Chen [EMAIL PROTECTED] wrote: Dear R users, If I have to combine plots which have the same independent and dependent variables in one graph. Which command should I use? Any example? Can I use panel? Viven, plot(x1,y1) par(new=TRUE) plot(x2,y2) You may have to use xlim and ylim to get the plots to match axes. Alternatively, plot(x1,y1) points(x2,y2) # or use lines, or whatnot Sean __ R-help@stat.math.ethz.ch 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] Combine related plots
You can also look at xyplot in the lattice package. You will have to set up your data slightly differently than for the standard graphics package, but it may well be worth learning to do so. The lattice package has enormous flexibility for combining multiple sets of data in one panel or plotting them in separate panels. Hope this helps, Matt -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Vivien W. Chen Sent: Wednesday, November 16, 2005 8:49 AM To: r-help@stat.math.ethz.ch Subject: [R] Combine related plots Dear R users, If I have to combine plots which have the same independent and dependent variables in one graph. Which command should I use? Any example? Can I use panel? Your help will be deeply appreciated. Thanks! -Vivien Chen- [[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] numericDeriv
(Ted Harding) [EMAIL PROTECTED] writes: On 16-Nov-05 Florent Bresson wrote: I have to compute some standard errors using the delta method and so have to use the command numericDeriv to get the desired gradient. Befor using it on my complicated function, I've done a try with a simple exemple : x - 1:5 numericDeriv(quote(x^2),x) and i get : [1] 1 8 27 64 125 216 attr(,gradient) [,1] [,2] [,3] [,4] [,5] [,6] [1,] Inf00 NaN00 [2,]000 NaN00 [3,]0 Inf0 NaN00 [4,]000 NaN00 [5,]00 Inf NaN00 [6,]000 NaN00 I don't understand the result. I thought I will get : [1] 1 8 27 64 125 216 attr(,gradient) [,1] [1,] 1 [2,] 4 [3,] 6 [4,] 8 [5,] 10 [6,] 12 The derivative of x^2 is still 2x, isn't it ? The trap you've fallen into is that x - 1:5 makes x of integer type, and (believe it or not) you cannot differentiate when the support of a function is the integers. Wrong topology (though I'm not sure that this is quite how R thinks about it). So give x a bit of elbow-room (numeric type has continous -- well, nearly -- topology): x - as.numeric(1:5) numericDeriv(quote(x^2),x) [1] 1 4 9 16 25 attr(,gradient) [,1] [,2] [,3] [,4] [,5] [1,]20000 [2,]04000 [3,]00600 [4,]00080 [5,]0000 10 Oho. That had me baffled for a while... We should probably also explain that x is a vector and differentiation of a vector w.r.t. a vector is a matrix. If you want a 5x1 result you should likely use something like d-0 numericDeriv(quote((x+d)^2),d) [1] 1 4 9 16 25 attr(,gradient) [,1] [1,]2 [2,]4 [3,]6 [4,]8 [5,] 10 (I *hope* there's no way to get the result that Florent claims that he expected...) -- 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] COM dates (was origin and origin- in chron)
I was just looking for an easy way to convert between COM datetime and chron datetime (both ways.) I found examples on the list, but they involved origin. Does anyone have functions for converting COM datetime - chron datetimethat work safely? David L. Reiner -Original Message- From: Gabor Grothendieck [mailto:[EMAIL PROTECTED] Sent: Tuesday, November 15, 2005 5:34 PM To: David Reiner [EMAIL PROTECTED] Cc: r-help@stat.math.ethz.ch Subject: [SPAM] - Re: [R] origin and origin- functions on chron - Bayesian Filter detected spam chron has a namespace so try this: chron:::origin getAnywhere(origin-) Having said that I would recommend that you don't use origins in chron. The situation may have changed but when I wrote the Help Desk article in R News 4/1 I encountered problems with using origins in some situations and as discussed in the article its so easy to avoid using them that there is really no good reason I can see not to avoid them. On 11/15/05, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: I'm trying to use/modify some code I found (at Omegahat, but I've seem similar usage elsewhere.) It contains the lines: if(any(origin(chronDate)!=orig)) origin(chronDate) - orig Let's say: require(chron) [1] TRUE chronDate - chron(11/15/2005, format=m/d/y, origin.=c(12,31,1899)) orig - c(month=12, day=31, year=1899) origin(chronDate) Error: couldn't find function origin origin(chronDate) - orig Error: couldn't find function origin- I'm sure I'm missing something simple here, but what? I've looked in the archives and docs quite a lot Thanks for the help! R-2.2.0 on Windows XP (SP2) R.Version() $platform [1] i386-pc-mingw32 $arch [1] i386 $os [1] mingw32 $system [1] i386, mingw32 $status [1] $major [1] 2 $minor [1] 2.0 $year [1] 2005 $month [1] 10 $day [1] 06 $svn rev [1] 35749 $language [1] R David L. Reiner Rho Trading 440 S. LaSalle St. Chicago IL 60605 312-362-4963 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting- guide.html David L. Reiner Rho Trading 440 S. LaSalle St. Chicago IL 60605 312-362-4963 [[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] COM dates (was origin and origin- in chron)
This is covered in the R News article. On 11/16/05, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: I was just looking for an easy way to convert between COM datetime and chron datetime (both ways.) I found examples on the list, but they involved origin. Does anyone have functions for converting COM datetime - chron datetimethat work safely? David L. Reiner -Original Message- From: Gabor Grothendieck [mailto:[EMAIL PROTECTED] Sent: Tuesday, November 15, 2005 5:34 PM To: David Reiner [EMAIL PROTECTED] Cc: r-help@stat.math.ethz.ch Subject: [SPAM] - Re: [R] origin and origin- functions on chron - Bayesian Filter detected spam chron has a namespace so try this: chron:::origin getAnywhere(origin-) Having said that I would recommend that you don't use origins in chron. The situation may have changed but when I wrote the Help Desk article in R News 4/1 I encountered problems with using origins in some situations and as discussed in the article its so easy to avoid using them that there is really no good reason I can see not to avoid them. On 11/15/05, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: I'm trying to use/modify some code I found (at Omegahat, but I've seem similar usage elsewhere.) It contains the lines: if(any(origin(chronDate)!=orig)) origin(chronDate) - orig Let's say: require(chron) [1] TRUE chronDate - chron(11/15/2005, format=m/d/y, origin.=c(12,31,1899)) orig - c(month=12, day=31, year=1899) origin(chronDate) Error: couldn't find function origin origin(chronDate) - orig Error: couldn't find function origin- I'm sure I'm missing something simple here, but what? I've looked in the archives and docs quite a lot Thanks for the help! R-2.2.0 on Windows XP (SP2) R.Version() $platform [1] i386-pc-mingw32 $arch [1] i386 $os [1] mingw32 $system [1] i386, mingw32 $status [1] $major [1] 2 $minor [1] 2.0 $year [1] 2005 $month [1] 10 $day [1] 06 $svn rev [1] 35749 $language [1] R David L. Reiner Rho Trading 440 S. LaSalle St. Chicago IL 60605 312-362-4963 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting- guide.html David L. Reiner Rho Trading 440 S. LaSalle St. Chicago IL 60605 312-362-4963 __ R-help@stat.math.ethz.ch 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] numericDeriv
Effectively, it's much better thanks --- Prof Brian Ripley [EMAIL PROTECTED] a écrit : On Wed, 16 Nov 2005, Florent Bresson wrote: I have to compute some standard errors using the delta method and so have to use the command numericDeriv to get the desired gradient. Befor using it on my complicated function, I've done a try with a simple exemple : x - 1:5 numericDeriv(quote(x^2),x) and i get : [1] 1 8 27 64 125 216 attr(,gradient) [,1] [,2] [,3] [,4] [,5] [,6] [1,] Inf00 NaN00 [2,]000 NaN00 [3,]0 Inf0 NaN00 [4,]000 NaN00 [5,]00 Inf NaN00 [6,]000 NaN00 I don't understand the result. I thought I will get : [1] 1 8 27 64 125 216 attr(,gradient) [,1] [1,] 1 [2,] 4 [3,] 6 [4,] 8 [5,] 10 [6,] 12 The derivative of x^2 is still 2x, isn't it ? and (1:5)^2 is still [1] 1 4 9 16 25 ! Try x - as.numeric(1:5) numericDeriv(quote(x^2),x) since the author of numericDeriv has forgotten some coercions. -- 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] x-axis in dendrogram
Hi there, Is it possible to fix the x-axis on a dendrogram and allow vertical lines to intersect to each other? Regards, Julio - [[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] Error in integrate
On Wed, 16 Nov 2005, Vivien W. Chen wrote: Hi! I am a beginner of R. I am trying to calculate integrate and draw a graph of the output, but just kept on getting error messages. I list my program and error message below. Please help. Many Thanks! === + a--11 b-0.1 c-0.012 x-0:110 This is harmless but suspicious -- this x is completely unrelated to the x in integrand() t-0:15 integrand-function(x) {exp(-exp(a-c*t)*(exp(b*x)-exp(c*x))/(b-c))} The problem is that x is a vector of whatever length integrate() decides to pass, but t is a vector of length 16. This mismatch of lengths is going to cause a problem. cal-integrate(integrand,0,Inf) Error in integrate(integrand, 0, Inf) : evaluation of function gave a result of wrong length In addition: Warning message: longer object length is not a multiple of shorter object length in: -exp(a - c * t) * (exp(b * x) - exp(c * x)) plot(cal~t, xlab=Time, ylab=Pop at t) Error in model.frame(formula, rownames, variables, varnames, extras, extranames, : invalid variable type === - Vivien Chen - [[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 Thomas Lumley Assoc. Professor, Biostatistics [EMAIL PROTECTED] University of Washington, Seattle __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] save to ascii
All, Usually when I save a variable I have ascii = FALSE since I usually load the variable via the load(variable) command and do not need to view the variable outside of R. When I set ascii = TRUE and view the variable outside of R in a text editor, I notice additional characters (starting w/ RDA2 ...) before the first actual value in the vector. Is there are way to save to ascii such that this does not happen? I checked the help under save and didn't see anything. I apologize in advance for this overly simplistic question. Cheers, Dave ps - please reply directly to [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
Re: [R] save to ascii
You could try using functions like 'dput()' or 'dump()'. These may or may not be equivalent to 'save()' depending on how complex the object to be saved is. -roger P.S. Asking people to respond to a different email address is not a good way to get responses! Afshartous, David wrote: All, Usually when I save a variable I have ascii = FALSE since I usually load the variable via the load(variable) command and do not need to view the variable outside of R. When I set ascii = TRUE and view the variable outside of R in a text editor, I notice additional characters (starting w/ RDA2 ...) before the first actual value in the vector. Is there are way to save to ascii such that this does not happen? I checked the help under save and didn't see anything. I apologize in advance for this overly simplistic question. Cheers, Dave ps - please reply directly to [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 -- Roger D. Peng | http://www.biostat.jhsph.edu/~rpeng/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] invert y-axis in barplot
Hi! This is probably a very trivial question. Is there an easy way to invert the y-axis (low values on top) when using the function barplot()? Thanks, Jörg [[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] Difficulties with for() {while(){}}
Em Qua 16 Nov 2005 11:30, jim holtman escreveu: What do you mean by 'abort'? Does an 'error' occur that you want to catch? If so, look at 'try'. Otherwise if it is testable, then just test for the condition and restart. Hi, this is not a real error, is a situation whitout a resolution. Is, I try to test the condition and restart, but how to restart all? Matthias suggest the use of initialize, I dont undertande how to use this. I try this: myfunction - function(...) { ## Init of function ... for(i in test) { ... while(j = test2) { ... test3 - make a test if(test3 == error) { myfunction(...) } } } } This is the best way to make this? In this case I need to put all arguments initialized in function(...) on the myfunction(...) . Thanks Ronaldo -- Faça algo à prova de idiotas e alguém fará um idiota melhor. -- | // | \\ [***] | ( õ õ ) [Ronaldo Reis Júnior] | V [UFV/DBA-Entomologia] |/ \ [36570-000 Viçosa - MG ] | /(.''`.)\ [Fone: 31-3899-4007 ] | /(: :' :)\ [EMAIL PROTECTED]] |/ (`. `'` ) \[ICQ#: 5692561 | LinuxUser#: 205366 ] |( `- ) [***] | _/ \_Powered by GNU/Debian Woody/Sarge __ R-help@stat.math.ethz.ch 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] save to ascii
Perhaps use dput? HTH, Andy From: Afshartous, David All, Usually when I save a variable I have ascii = FALSE since I usually load the variable via the load(variable) command and do not need to view the variable outside of R. When I set ascii = TRUE and view the variable outside of R in a text editor, I notice additional characters (starting w/ RDA2 ...) before the first actual value in the vector. Is there are way to save to ascii such that this does not happen? I checked the help under save and didn't see anything. I apologize in advance for this overly simplistic question. Cheers, Dave ps - please reply directly to [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] Warning message: package '...' was built under R version 2.3.0
Hi, While installing precompiled packages I often get warnings like the one in the subject. I usually ignore them, but I still do not understand why windows packages are build with unreleased versions of R. Is there some way to get packages build under R-2.2.0? What are potential problems that can result from that version mismatch? My system: winXP R version: 2.2.0 Jarek Tuszynski [[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] invert y-axis in barplot
On Wed, 2005-11-16 at 16:46 +, Jörg Schlingemann wrote: Hi! This is probably a very trivial question. Is there an easy way to invert the y-axis (low values on top) when using the function barplot()? Thanks, Jrg You mean something like this?: barplot(1:10, ylim = rev(c(0, 12))) or, something like this?: barplot(1:10, yaxt = n) axis(2, labels = rev(seq(0, 10, 2)), at = seq(0, 10, 2)) Note the use of rev() in each case. HTH, Marc Schwartz __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] nlme question
I am using the package nlme to fit a simple random effects (variance components model) with 3 parameters: overall mean (fixed effect), between subject variance (random) and within subject variance (random). I have 16 subjects with 1-4 obs per subject. I need a 3x3 variance-covariance matrix that includes all 3 parameters in order to compute the variance of a specific linear combination. But I can't get the 3x3 matrix. Should I specify the formulae in lme differently or is there some other suggestion that I might try? Thank you very much for any advice. my data and code follows. mydata - structure(list(subject = c(17, 17, 17, 17, 5, 16, 16, 8, 8, 8, 8, 7, 7, 7, 7, 9, 9, 9, 10, 10, 11, 11, 11, 12, 12, 12, 12, 14, 14, 14, 14, 15, 15, 15, 15, 13, 13, 13, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 4, 4, 4), y = c(-2.944, -5.521, -4.644, -4.736, -5.799, -4.635, -5.986, -5.011, -3.989, -4.682, -6.975, -6.064, -5.991, -8.068, -5.075, -5.298, -6.446, -5.037, -6.534, -4.828, -5.886, -4.025, -6.607, -5.914, -4.159, -6.757, -4.564, -5.011, -5.416, -5.371, -5.768, -7.962, -5.635, -4.575, -5.268, -6.975, -5.598, -7.669, -8.292, -7.265, -5.858, -7.003, -3.807, -5.829, -5.613, -3.135, -5.136, -5.394, -5.011, -5.598, -4.174)), .Names = c(subject, y), class = data.frame, row.names = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51)) lme.res-lme(fixed=y~1,data=mydata,random=~1|subject,method=ML) VarCorr(lme.res) [[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] invert y-axis in barplot
Jörg Schlingemann wrote: Hi! This is probably a very trivial question. Is there an easy way to invert the y-axis (low values on top) when using the function barplot()? barplot(1:10, ylim=c(10, 0)) works for me ... Uwe Ligges Thanks, Jörg [[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] Warning message: package '...' was built under R version 2.3.0
Tuszynski, Jaroslaw W. wrote: Hi, While installing precompiled packages I often get warnings like the one in the subject. I usually ignore them, but I still do not understand why windows packages are build with unreleased versions of R. Is there some way to get packages build under R-2.2.0? Well, those in the corresponding 2.2 repository on CRAN should all be build with R-2.2.0. If not, please tell me which one is wrong. Where did you get the packages for R-devel from? Uwe Ligges What are potential problems that can result from that version mismatch? My system: winXP R version: 2.2.0 Jarek Tuszynski [[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] Warning message: package '...' was built under R version 2.3.0
Uwe, I think I used following repositories: PA1, PA2 and NC, which are the closest to me and I think I got the same results in all. The package I downloaded was coin which downloaded other packages. The consol printout follows: chooseCRANmirror() utils:::menuInstallPkgs() trying URL 'http://www.ibiblio.org/pub/languages/R/CRAN/bin/windows/contrib/2.2/coin_0. 3-3.zip' Content type 'application/zip' length 793299 bytes opened URL downloaded 774Kb package 'coin' successfully unpacked and MD5 sums checked The downloaded packages are in C:\Documents and Settings\tuszynskij\Local Settings\Temp\Rtmp7594\downloaded_packages updating HTML package descriptions library(coin) Loading required package: survival Loading required package: splines Loading required package: mvtnorm Loading required package: modeltools Error in load(dataFile, ns) : ReadItem: unknown type 241 In addition: Warning messages: 1: package 'mvtnorm' was built under R version 2.3.0 2: package 'modeltools' was built under R version 2.3.0 Error: unable to load R code in package 'modeltools' Error: package 'modeltools' could not be loaded Jarek Tuszynski -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Wednesday, November 16, 2005 11:27 AM To: Tuszynski, Jaroslaw W. Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Warning message: package '...' was built under R version 2.3.0 Tuszynski, Jaroslaw W. wrote: Hi, While installing precompiled packages I often get warnings like the one in the subject. I usually ignore them, but I still do not understand why windows packages are build with unreleased versions of R. Is there some way to get packages build under R-2.2.0? Well, those in the corresponding 2.2 repository on CRAN should all be build with R-2.2.0. If not, please tell me which one is wrong. Where did you get the packages for R-devel from? Uwe Ligges What are potential problems that can result from that version mismatch? My system: winXP R version: 2.2.0 Jarek Tuszynski [[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] combination xyplot and barchart?
Thank you, that worked great. Per your suggestion, I transformed the Precip data to fit on the same scale as the Temp data. After changing trellis.par.set$layout.widths$right.padding to 5, I was able to fit a second axis and label for the Precip (using panel.axis) to achieve the plot I want. Erin On Nov 15, 2005, at 9:09 PM, Deepayan Sarkar wrote: On 11/15/05, Erin Berryman [EMAIL PROTECTED] wrote: Dear R community, I am having trouble determining how to create the graph I want utilizing my relatively limited knowledge of R. So far I have been using the lattice library to create most of what I need. The dataset (enviro) consists of 2 variables (Temp and Precip) for each Day of a 2-yr period (Year). I wish to display Temp and Precip along the y axis plotted by Day on the x axis to allow comparison (one year's data in each of 2 panels stacked on top of each other) between the years. Essentially what I want it to look like is an xyplot (Temp ~ Day | Year, type='l') superimposed onto a barchart(Precip ~ Day | Year, horizontal=F), with scales adjusted so one can see detail in both variables. The closest I have come to what I need is by the following code : library(lattice) barchart(Precip + Temp ~ Day | Year, data=enviro, layout=c(1,2), horizontal=F, origin=0, panel=function(x,y,subscripts,...){panel.xyplot(x=enviro$Day, y=enviro$Temp, type='l',subscripts=subscripts, ...); panel.barchart(x=enviro$Day, y=enviro$Precip, subscripts=subscripts, ...)}) Two panels are produced; however, both years' data are plotted in each panel (panels look identical). And I get this error: Error in grid.Call.graphics(L_rect, x$x, x$y, x$width, x$height, resolveHJust(x$just, : invalid line type A reproducible example, even if it's a toy one, would have been helpful. Your usage is confused. In particular, panel.xyplot ignores the subscripts argument, so you end up giving exactly the same set of values to panel.xyplot for both panels (so it's not surprising that your panels show the same data). It seems that you are looking for something like the following: enviro - data.frame(Year = rep(2001:2002, each = 365), Day = rep(1:365, 2), Precip = pmax(0, rnorm(365 * 2)), Temp = 2 + 0.2 * rnorm(365 * 2)) xyplot(Precip + Temp ~ Day | Year, data=enviro, layout = c(1, 2), panel = panel.superpose.2, type = c('h', 'l')) In case it helps, this is shorthand for xyplot(Precip + Temp ~ Day | Year, data=enviro, layout = c(1, 2), panel = function(x, y, groups, subscripts, ...) { panel.superpose.2(x = x, y = y, groups = groups, subscripts = subscripts, ...) }, type = c('h', 'l')) Note that the panel function is defined in terms of arguments it gets, and does not explicitly refer to any external variables (like the data frame 'enviro'). If you find yourself writing code that does, it's a likely sign that you are doing something wrong (or at least unnecessarily convoluted). From the documentation or the help archives, I cannot understand how to: 1) indicate a conditioning variable (Year) for panel.barchart and panel.xyplot 2) have 2 y axes with different scales in one panel You can't easily. A panel has one set of scales, and that's it. You can of course fake it by transforming the relevant part of your data and adding a set of tick marks with appropriate (fake) labels (see ?panel.axis). -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
[R] nmle question
Hello. I have 16 subjects with 1-4 obs per subject. I am using the package nlme to fit a simple random effects (variance components model) with 3 parameters: overall mean (fixed effect), between subject variance (random) and within subject variance (random). I need a 3x3 variance-covariance matrix that includes all 3 parameters in order to compute the variance of a specific linear combination. But I can't get the 3x3 matrix. Should I specify the formulae in lme differently or is there some other suggestion that I might try? Thank you very much for any advice. my data and code follows. mydata - structure(list(subject = c(17, 17, 17, 17, 5, 16, 16, 8, 8, 8, 8, 7, 7, 7, 7, 9, 9, 9, 10, 10, 11, 11, 11, 12, 12, 12, 12, 14, 14, 14, 14, 15, 15, 15, 15, 13, 13, 13, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 4, 4, 4), y = c(-2.944, -5.521, -4.644, -4.736, -5.799, -4.635, -5.986, -5.011, -3.989, -4.682, -6.975, -6.064, -5.991, -8.068, -5.075, -5.298, -6.446, -5.037, -6.534, -4.828, -5.886, -4.025, -6.607, -5.914, -4.159, -6.757, -4.564, -5.011, -5.416, -5.371, -5.768, -7.962, -5.635, -4.575, -5.268, -6.975, -5.598, -7.669, -8.292, -7.265, -5.858, -7.003, -3.807, -5.829, -5.613, -3.135, -5.136, -5.394, -5.011, -5.598, -4.174)), .Names = c(subject, y), class = data.frame, row.names = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51)) lme.res-lme(fixed=y~1,data=mydata,random=~1|subject,method=ML) VarCorr(lme.res) __ R-help@stat.math.ethz.ch 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] Unexpected result of names-
That's because S does partial matching of component names, if it can be done unambiguously: z - list(aaa=1, bbb=2) z$a [1] 1 z[[a]] [1] 1 z[a] $NA NULL z[aaa] $aaa [1] 1 Andy From: Hong Ooi Hi, I came across some rather unexpected behaviour the other day with assigning names and lists. Here's an example. z - list(aaa=1, bbb=2) z $aaa [1] 1 $bbb [1] 2 Note that z has members named aaa and bbb. Now: names(z$a) - X z $aaa [1] 1 $bbb [1] 2 $a X 1 I would have expected that trying to name z$a would either give an error (because z doesn't have an element a) or would cause z$aaa to be modified (due to partial name matching). I didn't expect that a new list member would be created. I've checked that this is consistent across SPlus 2000, SPlus 7, and R 2.2 for Windows. Can someone give an explanation for why this is happening? -- Hong Ooi Senior Research Analyst, IAG Limited 388 George St, Sydney NSW 2000 (02) 9292 1566 __ _ The information transmitted in this message and its attachments (if any) is intended only for the person or entity to which it is addressed. The message may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon this information, by persons or entities other than the intended recipient is prohibited. If you have received this in error, please contact the sender and delete this e-mail and associated material from any computer. The intended recipient of this e-mail may only use, reproduce, disclose or distribute the information contained in this e-mail and any attached files, with the permission of the sender. This message has been scanned for viruses with Symantec Scan Engine and cleared by MailMarshal. __ R-help@stat.math.ethz.ch 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] Newton-Raphson
Dear all, I want to solve a score function by using Newton-Raphson algorithm. Is there such a fucntion in R? I know there's one called optim, but it seems only doing minimizing or maximizing. Thanks, Jimmy __ R-help@stat.math.ethz.ch 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] RODBC help
I am using the RODBC package to read data from an Excel file. An excerpt of the file looks like this: 00103V206 AES Corporation 6.00% 42.87 00808N202 AES Trust III 6.75% 34.98 03748R861 Apartment Investment Management 9.00% #ERROR (I) 039380209 Arch Coal, Inc. 5.00% 61.51 My problem appears to be related to specification of data types by column. For instance, the CUSIPS in column 1 are read in correctly when they contain a letter, but as NA when they are purely numbers. Similarly, in the fourth column the rows with #ERROR (I) are read in as such, but all other values become NA. I have experimented with as.is but have not been able to arrive at a solution. I also changed the nullstrings return value and each of the NAs generated above appear to be genrated by the query seeing null strings. As always, your assistance is most appreciated. _ Dont just search. Find. Check out the new MSN Search! __ R-help@stat.math.ethz.ch 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] Warning message: package '...' was built under R version 2.3.0
Tuszynski, Jaroslaw W. wrote: Uwe, I think I used following repositories: PA1, PA2 and NC, which are the closest to me and I think I got the same results in all. The package I downloaded was coin which downloaded other packages. The consol printout follows: chooseCRANmirror() utils:::menuInstallPkgs() trying URL 'http://www.ibiblio.org/pub/languages/R/CRAN/bin/windows/contrib/2.2/coin_0. 3-3.zip' Content type 'application/zip' length 793299 bytes opened URL downloaded 774Kb package 'coin' successfully unpacked and MD5 sums checked The downloaded packages are in C:\Documents and Settings\tuszynskij\Local Settings\Temp\Rtmp7594\downloaded_packages updating HTML package descriptions library(coin) Loading required package: survival Loading required package: splines Loading required package: mvtnorm Loading required package: modeltools Error in load(dataFile, ns) : ReadItem: unknown type 241 In addition: Warning messages: 1: package 'mvtnorm' was built under R version 2.3.0 I have just downloaded http://www.ibiblio.org/pub/languages/R/CRAN/bin/windows/contrib/2.2/mvtnorm_0.7-2.zip and its DESCRIPTION file correctly has: Built: R 2.2.0; i386-pc-mingw32; 2005-10-14 14:44:12; windows My guess is that you are using one library for two different R versions and you have installed some R-devel package into the library you are now using with R-2.2.0 ... You cannot mix binary packages for R 2.3.0 with those for R-devel due to changes in environment handling. Uwe Ligges 2: package 'modeltools' was built under R version 2.3.0 Error: unable to load R code in package 'modeltools' Error: package 'modeltools' could not be loaded Jarek Tuszynski -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Wednesday, November 16, 2005 11:27 AM To: Tuszynski, Jaroslaw W. Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Warning message: package '...' was built under R version 2.3.0 Tuszynski, Jaroslaw W. wrote: Hi, While installing precompiled packages I often get warnings like the one in the subject. I usually ignore them, but I still do not understand why windows packages are build with unreleased versions of R. Is there some way to get packages build under R-2.2.0? Well, those in the corresponding 2.2 repository on CRAN should all be build with R-2.2.0. If not, please tell me which one is wrong. Where did you get the packages for R-devel from? Uwe Ligges What are potential problems that can result from that version mismatch? My system: winXP R version: 2.2.0 Jarek Tuszynski [[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] normal cdf over an interval
Hi, I'm trying to find a way to take evaluate the Normal CDF over an interval and return the result on the log scale. This works, but I think it isn't numerically stable: log(pnorm(a, mean = x, sd = y) - pnorm(b, mean = x, sd = y)) Does anyone know of a single function that does the above? Or knows of a way to make it more stable? I'd really appreciate any suggestions! Thanks, Olivia [[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] RODBC and Very long field lengths
I'm having a small problem using RODBC. I'm trying to retrieve a string from a very long memo field (512*20*9=9360 characters = 74880 bytes) in an MSAccess database. It appears that RODBC set a maximum buffer size for a single column of 65535 bytes. ## cut from RODBC.c ## } else { /* transfer as character */ int datalen = thisHandle-ColData[i].ColSize; if (datalen = 0 || datalen COLMAX) datalen = COLMAX; /* sanity check as the reports are sometimes unreliable */ if (datalen 65535) datalen = 65535; ## Can I increase this by just changeing the value in RODBC.c? If so how do I get R to re-compile the package? Thanks for your advice Tom __ R-help@stat.math.ethz.ch 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] Unexpected result of names-
On Wed, 16 Nov 2005, Liaw, Andy wrote: That's because S does partial matching of component names, if it can be done unambiguously: z - list(aaa=1, bbb=2) z$a [1] 1 z[[a]] [1] 1 z[a] $NA NULL [That could be done unambiguously, but [] does not partial match in R (see ?[), only [[]] and $ do. That is different from S (Blue Book p.358), and I do not know why.] z[aaa] $aaa [1] 1 Well, *sometimes*. For z - list(aaa=1, bbb=2) z$a - 3 z $aaa [1] 1 $bbb [1] 2 $a [1] 3 It is that the rules are different for extraction and replacement that surprises people. In this case we have effectively z - $-(names-(z$a, X), a) and the replacement function $- does not partially match (Blue Book, p.362). Andy From: Hong Ooi Hi, I came across some rather unexpected behaviour the other day with assigning names and lists. Here's an example. z - list(aaa=1, bbb=2) z $aaa [1] 1 $bbb [1] 2 Note that z has members named aaa and bbb. Now: names(z$a) - X z $aaa [1] 1 $bbb [1] 2 $a X 1 I would have expected that trying to name z$a would either give an error (because z doesn't have an element a) or would cause z$aaa to be modified (due to partial name matching). I didn't expect that a new list member would be created. I've checked that this is consistent across SPlus 2000, SPlus 7, and R 2.2 for Windows. Can someone give an explanation for why this is happening? -- Hong Ooi Senior Research Analyst, IAG Limited 388 George St, Sydney NSW 2000 (02) 9292 1566 __ _ The information transmitted in this message and its attachments (if any) is intended only for the person or entity to which it is addressed. The message may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon this information, by persons or entities other than the intended recipient is prohibited. If you have received this in error, please contact the sender and delete this e-mail and associated material from any computer. The intended recipient of this e-mail may only use, reproduce, disclose or distribute the information contained in this e-mail and any attached files, with the permission of the sender. This message has been scanned for viruses with Symantec Scan Engine and cleared by MailMarshal. __ R-help@stat.math.ethz.ch 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 -- 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] Warning message: package '...' was built under R version 2.3.0
I do have single version of R (2.2.0) and all packages I have were installed by downloading precompiled packages from PA or NC repositories within last month. I also just managed to download 'modeltools' from PA2 and its DESCRIPTION says: Built: R 2.3.0; ; 2005-11-04 20:47:47; windows R console: utils:::menuInstallPkgs() --- Please select a CRAN mirror for use in this session --- trying URL 'http://lib.stat.cmu.edu/R/CRAN/bin/windows/contrib/2.2/modeltools_0.2-0.zip ' Content type 'application/zip' length 64426 bytes opened URL downloaded 62Kb package 'modeltools' successfully unpacked and MD5 sums checked The downloaded packages are in C:\Documents and Settings\tuszynskij\Local Settings\Temp\Rtmp4458\downloaded_packages updating HTML package descriptions library(modeltools) Error in load(dataFile, ns) : ReadItem: unknown type 241 In addition: Warning message: package 'modeltools' was built under R version 2.3.0 Error: unable to load R code in package 'modeltools' Error: package/namespace load failed for 'modeltools' The same package downloaded from PA1 does not give me any warnings or errors. Jarek Tuszynski -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Wednesday, November 16, 2005 1:04 PM To: Tuszynski, Jaroslaw W. Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Warning message: package '...' was built under R version 2.3.0 Tuszynski, Jaroslaw W. wrote: Uwe, I think I used following repositories: PA1, PA2 and NC, which are the closest to me and I think I got the same results in all. The package I downloaded was coin which downloaded other packages. The consol printout follows: chooseCRANmirror() utils:::menuInstallPkgs() trying URL 'http://www.ibiblio.org/pub/languages/R/CRAN/bin/windows/contrib/2.2/c oin_0. 3-3.zip' Content type 'application/zip' length 793299 bytes opened URL downloaded 774Kb package 'coin' successfully unpacked and MD5 sums checked The downloaded packages are in C:\Documents and Settings\tuszynskij\Local Settings\Temp\Rtmp7594\downloaded_packages updating HTML package descriptions library(coin) Loading required package: survival Loading required package: splines Loading required package: mvtnorm Loading required package: modeltools Error in load(dataFile, ns) : ReadItem: unknown type 241 In addition: Warning messages: 1: package 'mvtnorm' was built under R version 2.3.0 I have just downloaded http://www.ibiblio.org/pub/languages/R/CRAN/bin/windows/contrib/2.2/mvtnorm_ 0.7-2.zip and its DESCRIPTION file correctly has: Built: R 2.2.0; i386-pc-mingw32; 2005-10-14 14:44:12; windows My guess is that you are using one library for two different R versions and you have installed some R-devel package into the library you are now using with R-2.2.0 ... You cannot mix binary packages for R 2.3.0 with those for R-devel due to changes in environment handling. Uwe Ligges __ R-help@stat.math.ethz.ch 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] normal cdf over an interval
On 16-Nov-05 Olivia Lau wrote: Hi, I'm trying to find a way to take evaluate the Normal CDF over an interval and return the result on the log scale. This works, but I think it isn't numerically stable: log(pnorm(a, mean = x, sd = y) - pnorm(b, mean = x, sd = y)) Does anyone know of a single function that does the above? Or knows of a way to make it more stable? I'd really appreciate any suggestions! Thanks, Olivia A little more detail might be helpful. Your formula is simple enough in itself, and you should not be getting numerical stability problems for reasonable values of (a-x)/y or (b-x)/y (say in the range -4 to 4). And I assume you're being careful that a b! So for what values of your variables are problems arising? And how do they manifest themselves? If it's due to large values of (a-x)/y etc., then possibly an asymptotic approximation may serve. There are some good ones around. Can you tell us a bit more? Best wishes, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 16-Nov-05 Time: 20:37:47 -- 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] RODBC and Very long field lengths
On Wed, 16 Nov 2005, tom wright wrote: I'm having a small problem using RODBC. I'm trying to retrieve a string from a very long memo field (512*20*9=9360 characters = 74880 bytes) in an MSAccess database. It appears that RODBC set a maximum buffer size for a single column of 65535 bytes. Well, it is documented in the ChangeLog. (The limit used to be much lower.) ## cut from RODBC.c ## } else { /* transfer as character */ int datalen = thisHandle-ColData[i].ColSize; if (datalen = 0 || datalen COLMAX) datalen = COLMAX; /* sanity check as the reports are sometimes unreliable */ if (datalen 65535) datalen = 65535; ## Can I increase this by just changeing the value in RODBC.c? If so how do I get R to re-compile the package? Yes, and you re-compile it just like any other package, see the rw-FAQ, the R-admin manual and the latest Helpdesk in R-news, for example. -- 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] Warning message: package '...' was built under R version 2.3.0
Tuszynski, Jaroslaw W. [EMAIL PROTECTED] writes: I do have single version of R (2.2.0) and all packages I have were installed by downloading precompiled packages from PA or NC repositories within last month. I also just managed to download 'modeltools' from PA2 and its DESCRIPTION says: Built: R 2.3.0; ; 2005-11-04 20:47:47; windows R console: utils:::menuInstallPkgs() --- Please select a CRAN mirror for use in this session --- trying URL 'http://lib.stat.cmu.edu/R/CRAN/bin/windows/contrib/2.2/modeltools_0.2-0.zip ' Content type 'application/zip' length 64426 bytes opened URL downloaded 62Kb package 'modeltools' successfully unpacked and MD5 sums checked The downloaded packages are in C:\Documents and Settings\tuszynskij\Local Settings\Temp\Rtmp4458\downloaded_packages updating HTML package descriptions library(modeltools) Error in load(dataFile, ns) : ReadItem: unknown type 241 In addition: Warning message: package 'modeltools' was built under R version 2.3.0 Error: unable to load R code in package 'modeltools' Error: package/namespace load failed for 'modeltools' The same package downloaded from PA1 does not give me any warnings or errors. Jarek Tuszynski Yes. The Statlib mirror (which I suppose is what you call PA2) appears to be badly messed up. The packages under http://lib.stat.cmu.edu/R/CRAN/bin/windows/contrib seem not to have been updated since 2004, whereas (e.g.) http://lib.stat.cmu.edu/R/CRAN/bin/windows/contrib/r-release/gRbase_0.1.23.zip carries the Built: 2.3.0 line in the DESCRIPTION file. Even stranger, looking in R/CRAN/bin/windows/contrib shows no subfolder r-release, and R/CRAN/bin/windows/contrib/r-release contains version 0.1.13 of gRbase! In short, use another mirror until this one gets sorted out... -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Wednesday, November 16, 2005 1:04 PM To: Tuszynski, Jaroslaw W. Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Warning message: package '...' was built under R version 2.3.0 Tuszynski, Jaroslaw W. wrote: Uwe, I think I used following repositories: PA1, PA2 and NC, which are the closest to me and I think I got the same results in all. The package I downloaded was coin which downloaded other packages. The consol printout follows: chooseCRANmirror() utils:::menuInstallPkgs() trying URL 'http://www.ibiblio.org/pub/languages/R/CRAN/bin/windows/contrib/2.2/c oin_0. 3-3.zip' Content type 'application/zip' length 793299 bytes opened URL downloaded 774Kb package 'coin' successfully unpacked and MD5 sums checked The downloaded packages are in C:\Documents and Settings\tuszynskij\Local Settings\Temp\Rtmp7594\downloaded_packages updating HTML package descriptions library(coin) Loading required package: survival Loading required package: splines Loading required package: mvtnorm Loading required package: modeltools Error in load(dataFile, ns) : ReadItem: unknown type 241 In addition: Warning messages: 1: package 'mvtnorm' was built under R version 2.3.0 I have just downloaded http://www.ibiblio.org/pub/languages/R/CRAN/bin/windows/contrib/2.2/mvtnorm_ 0.7-2.zip and its DESCRIPTION file correctly has: Built: R 2.2.0; i386-pc-mingw32; 2005-10-14 14:44:12; windows My guess is that you are using one library for two different R versions and you have installed some R-devel package into the library you are now using with R-2.2.0 ... You cannot mix binary packages for R 2.3.0 with those for R-devel due to changes in environment handling. Uwe Ligges __ R-help@stat.math.ethz.ch 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] X11 error in png
Hi all, When I ran a script containing the following codes: png(paste(savepath,a_rnaplot.png,sep = ),width = fwidth, height = fheight,pointsize = fpointsize); data_deg - AffyRNAdeg(data_cel) plotAffyRNAdeg(data_deg,col=cols,lty=1,lwd = 2) #a - par(fin) legend(bottomright,sampleNames(data_cel),col=cols,lty=1) RNAdegSlope = cbind(Sample Names = data_deg$sample.names, RNA Deg Slope = data_deg$slope) write.csv(RNAdegSlope, file = paste(savepath,RNAdegSlope.csv,sep = )) dev.off() I got this error message: Error in X11(paste(png::, filename, sep = ), width, height, pointsize, : unable to start device PNG In addition: Warning message: unable to open connection to X11 display '' My system: Linux AS 4.0 64-bit PATH=/usr/kerberos/sbin:/usr/kerberos/bin:/usr/local/bin:/bin:/usr/bin:/usr/X11 Jack Zhu, MD, MS Lombardi Cancer center Georgetown University 3970 Reservoir Rd, NW, NRB, W405b Washington, DC 20057 Email: [EMAIL PROTECTED] Tel: (202)-687-7451 Web: http://clarkelabs.georgetown.edu [[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] How to choose a validation set
Dear R-helpers, I am working with a time series data set of 1000 observations and I am runnig several models on it. I divided the sample in three sets: estimation [1:700], validation, [701:990] and test [991:1000]. Now I was wondering if there is a more elegant way to do this. Are you aware of a standard approach using R functions? Or if you could just suggest me some references I would really appreaciate it. Thansk a lot in advance. Regards, J. - [[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] X11 error in png
Does this help? http://cran.r-project.org/doc/FAQ/R-FAQ.html#How-do-I-produce-PNG-graphics-i n-batch-mode_003f Andy From: Jack Zhu Hi all, When I ran a script containing the following codes: png(paste(savepath,a_rnaplot.png,sep = ),width = fwidth, height = fheight,pointsize = fpointsize); data_deg - AffyRNAdeg(data_cel) plotAffyRNAdeg(data_deg,col=cols,lty=1,lwd = 2) #a - par(fin) legend(bottomright,sampleNames(data_cel),col=cols,lty=1) RNAdegSlope = cbind(Sample Names = data_deg$sample.names, RNA Deg Slope = data_deg$slope) write.csv(RNAdegSlope, file = paste(savepath,RNAdegSlope.csv,sep = )) dev.off() I got this error message: Error in X11(paste(png::, filename, sep = ), width, height, pointsize, : unable to start device PNG In addition: Warning message: unable to open connection to X11 display '' My system: Linux AS 4.0 64-bit PATH=/usr/kerberos/sbin:/usr/kerberos/bin:/usr/local/bin:/bin: /usr/bin:/usr/X11 Jack Zhu, MD, MS Lombardi Cancer center Georgetown University 3970 Reservoir Rd, NW, NRB, W405b Washington, DC 20057 Email: [EMAIL PROTECTED] Tel: (202)-687-7451 Web: http://clarkelabs.georgetown.edu [[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] Warning message: package '...' was built under R version 2.3.0
Yes. The Statlib mirror (which I suppose is what you call PA2) appears to be badly messed up. -- 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 Peter, Thanks for checking on this problem. My names for CRAN mirror sites (PA 1, PA 2, NC, etc.) came from that is shown in windows RGui when you choose Packages/Set CRAN mirror menu. All mirrors are represented by Country (city) for most of the world except US which is shown as USA (state abbreviation + number). So what I call PA 1 means USA/Pennsylvania #1, etc. Names you are using (Statlib mirror) are not visible in GUI. Just a clarification. Jarek Tuszynski __ R-help@stat.math.ethz.ch 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] update R packages in local repos
I try to update R packages via my local repository. I put all R packages in g:/myFolder/myRepository, I do like library(tools) write_PACKAGES(g:/myFolder/myRepository) options(repos=c(LocalR=file://g:/myFolder/myRepository)) getOption(repos) LocalR file://g:/myFolder/myRepository update.packages(ask = graphics) Error in gzfile(file, r) : unable to open connection In addition: Warning message: cannot open compressed file ':/myFolder/myRepository/bin/windows/contrib/2.2/PACKAGES' ?update.packages It produces these file in g:/myFolder/myRepository PACKAGES PACKAGES.gz Could I make this folder (bin/windows/contrib/2.2/)? Why? Regards, Muhammad Subianto version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major2 minor2.0 year 2005 month10 day 06 svn rev 35749 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] update R packages in local repos
This should work: update.packages(ask = graphics, repos = NULL, contriburl = file:///g:/myFolder/myRepository)) -peter Muhammad Subianto wrote: I try to update R packages via my local repository. I put all R packages in g:/myFolder/myRepository, I do like library(tools) write_PACKAGES(g:/myFolder/myRepository) options(repos=c(LocalR=file://g:/myFolder/myRepository)) getOption(repos) LocalR file://g:/myFolder/myRepository update.packages(ask = graphics) Error in gzfile(file, r) : unable to open connection In addition: Warning message: cannot open compressed file ':/myFolder/myRepository/bin/windows/contrib/2.2/PACKAGES' ?update.packages It produces these file in g:/myFolder/myRepository PACKAGES PACKAGES.gz Could I make this folder (bin/windows/contrib/2.2/)? Why? Regards, Muhammad Subianto version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major2 minor2.0 year 2005 month10 day 06 svn rev 35749 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 -- Peter Ehlers Department of Mathematics and Statistics University of Calgary, 2500 University Dr. NW __ R-help@stat.math.ethz.ch 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] X11 error in png
I managed to delete the original post about this and will no doubt mess up threading of responses. The post is shown below. This looks like a problem I saw just a couple of days ago. I noticed that you used a path variable in determining the file location and name. If something is wrong with your path (e.g. the directories it specifies don't exist) I think you will get this error. Check your path to make sure it is ok. David Scott *** Original post *** From: Jack Zhu yz8_at_georgetown.edu Date: Thu 17 Nov 2005 - 08:47:27 EST Hi all, When I ran a script containing the following codes: png(paste(savepath,a_rnaplot.png,sep = ),width = fwidth, height = fheight,pointsize = fpointsize); data_deg - AffyRNAdeg(data_cel) plotAffyRNAdeg(data_deg,col=cols,lty=1,lwd = 2) #a - par(fin) legend(bottomright,sampleNames(data_cel),col=cols,lty=1) RNAdegSlope = cbind(Sample Names = data_deg$sample.names, RNA Deg Slope = data_deg$slope) write.csv(RNAdegSlope, file = paste(savepath,RNAdegSlope.csv,sep = )) dev.off() I got this error message: Error in X11(paste(png::, filename, sep = ), width, height, pointsize, : unable to start device PNG In addition: Warning message: unable to open connection to X11 display '' My system: Linux AS 4.0 64-bit PATH=/usr/kerberos/sbin:/usr/kerberos/bin:/usr/local/bin:/bin:/usr/bin:/usr/X11 Jack Zhu, MD, MS Lombardi Cancer center Georgetown University 3970 Reservoir Rd, NW, NRB, W405b Washington, DC 20057 Email: [EMAIL PROTECTED] Tel: (202)-687-7451 Web: http://clarkelabs.georgetown.edu -- _ David Scott Department of Statistics, Tamaki Campus The University of Auckland, PB 92019 AucklandNEW ZEALAND Phone: +64 9 373 7599 ext 86830 Fax: +64 9 373 7000 Email: [EMAIL PROTECTED] Graduate Officer, Department of Statistics __ R-help@stat.math.ethz.ch 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] GEE in Fortran
Dear all, Does anybody happen to have written some GEE source code in Fortran to solve a Generalized Estimation Equation Model? Or kindly point out me a good starting point. Thanks. Zhu Wang SCHARP __ R-help@stat.math.ethz.ch 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] Is it feasible with R?
My data are (for one experiment): 3 Types of plant (1,2,3) 4 Species per Type (ex:for type 1(a,b,c,d), for type 2(e,f,g,h) and for type 3(i,j,k,l) 8 Repetitions of each Species 3 Stages (10, 20, 30) 2 Measures per Stages, for the stages 20 and 30 (ex: for Stage 10(C), for Stage 20 (A and B) and for Stage 30 (A and B) 3 Types x 4 Species(Type) x 8 Repetitions x (1 Stage x 1 Mesure + 2 Stages x 2 Measures= 480 data. There are 10 data missing. The final number of data is 470. My questions that I ask to my data: 1- Is there a significant difference between Measure A and Measure B? 2- Is there a significant difference between Stages 10, 20 and 30? 3- Is there a significant difference between the 3 plant Types? I haven't found any way to consider the whole data lot in one shot, I decided to remove the first stage with the Measure C that comes with it. It bring the number of data to 384. Is it possible to consider the whole lot in one shot? Thus my model is Sourced.l. Types 2 Species (Type) 9 Error a)84 Stage 1 Measure 1 Stage x Measure 1 Stage x Type2 Measure x Type 2 Stage x Measure x Type 2 Stage x Species (Type) 9 Measure x Species (Type) 9 Stage x Measure x Species (Type)9 Erreur b) 252 Total 383 How can I write this in R? My problem is that when I use %in% to nest Species in Type, R looks, for exemple, for specie a in Type 2 and it finds nothing and returns Na. Type1:Plante[T.a] -1.519670.82774 -1.836 0.0672 . Type2:Plante[T.a]NA NA NA NA Can you help me? Can R consider Species a,b,c,d only in Type 1, Species e,f,g,h only in Type 2 and Species i,j,k,l only in Type 3? Thank you very much. __ R-help@stat.math.ethz.ch 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] Portable R?
Hello list, A short time ago, I found http://johnhaller.com/jh/useful_stuff/portable_apps_suite/, which contains basically a complete set of office tools that can be run *entirely* from a USB key. The concept is: - find a Windows PC - put in your USB key - run OpenOffice, Firefox, Gaim, Nvu, Thunderbird, ... directly from your USB key, with no app installation required - save your files wherever - remove your USB key and leave, with nothing installed on the original PC As a consultant who battles regularly with limited toolsets at customer sites, this strikes me as an extremely handy way of working. Has anyone managed to setup a base R configuration that runs entirely from USB key? Being a regular user, but no expert, with R, it'd be very helpful for me if such a mechanism existed, but I've got no idea where to begin in building such a thing. Thanks in advance for any responses or suggestions Dave M. __ R-help@stat.math.ethz.ch 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] Portable R?
David, I've run old version of R (2 years ago) from a CD and it worked well for me. I guess you could run R from a USB drive. On 11/16/05, David Mitchell [EMAIL PROTECTED] wrote: Hello list, A short time ago, I found http://johnhaller.com/jh/useful_stuff/portable_apps_suite/, which contains basically a complete set of office tools that can be run *entirely* from a USB key. The concept is: - find a Windows PC - put in your USB key - run OpenOffice, Firefox, Gaim, Nvu, Thunderbird, ... directly from your USB key, with no app installation required - save your files wherever - remove your USB key and leave, with nothing installed on the original PC As a consultant who battles regularly with limited toolsets at customer sites, this strikes me as an extremely handy way of working. Has anyone managed to setup a base R configuration that runs entirely from USB key? Being a regular user, but no expert, with R, it'd be very helpful for me if such a mechanism existed, but I've got no idea where to begin in building such a thing. Thanks in advance for any responses or suggestions Dave M. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- WenSui Liu (http://statcompute.blogspot.com) Senior Decision Support Analyst Cincinnati Children Hospital Medical Center [[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] Portable R?
On 17 November 2005 at 14:16, David Mitchell wrote: | Hello list, | | A short time ago, I found | http://johnhaller.com/jh/useful_stuff/portable_apps_suite/, which | contains basically a complete set of office tools that can be run | *entirely* from a USB key. The concept is: | - find a Windows PC | - put in your USB key | - run OpenOffice, Firefox, Gaim, Nvu, Thunderbird, ... directly from | your USB key, with no app installation required | - save your files wherever | - remove your USB key and leave, with nothing installed on the original PC | | As a consultant who battles regularly with limited toolsets at | customer sites, this strikes me as an extremely handy way of working. | | Has anyone managed to setup a base R configuration that runs entirely | from USB key? Being a regular user, but no expert, with R, it'd be | very helpful for me if such a mechanism existed, but I've got no idea | where to begin in building such a thing. Short answer: Yes but using Linux, requiring a larger USB stick and some fiddling. Longer answer: Quantian (http://dirk.eddelbuettel.com/quantian) is a everything, the kitchen sink and some Linux distribution running off a DVD. Quantian is focussed on quantitative / numeric apps, and tends to include R plus related goodies -- the last release had an almost complete set of CRAN and BioConductor packages. The raw size of the last release is around 2 GB, corresponding to 6.6 GB expanded. Marco Caliari, who often contributes improved boot code to Quantian, has managed to boot Quantian off a USB stick. I didn't manage to do that with my laptop, possibly because of limitations in its bios. Some of this was discussed in past threads on the quantian-general mailing list. Lots-o-work suggestion: To not require a huge USB stick, you could try to shrink a given live cdrom such as Knoppix or Ubuntu, then add R and other goodies such that you're left with around 512 MB compressed. Then throw it onto a USB stick and make it bootable. Shortcut: Order a Quantian DVD. Some folks sell them pre-made for less than $5. Experiment with that, If you like it, consider making your own mini-distro. Or stick with the DVD and use it directly with the USB stick for your configuration, data, demos, ... Even shorter: R is perfectly relocatable. If you install the Windows binary onto the USB drive, it will run fine. You'll probably need to add editors and other tools. Hope this helps, Dirk -- Statistics: The (futile) attempt to offer certainty about uncertainty. -- Roger Koenker, 'Dictionary of Received Ideas of Statistics' __ R-help@stat.math.ethz.ch 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] some questions
Dear R expert, The following is my questions: 1. How to generate two sequences of 150 uniform deviates, called v1 and v2, in the range [-1,1]. 2. How to compute r=(v1)^2+(v2)^2 3.If r is outside the range of (0,1) then it will be discarded. 4 How to compute (v1)*sqrt(-2*log(r)/r) and output the first 100 z values in the list. 5. Plot histograms for the sequences from steps 1 and 2. Thanks very much for your help! xingyu __ R-help@stat.math.ethz.ch 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] R questions
Dear Sir/Madam, I am a beginner in R. Here is my questions. 1. Can you give me one test for randomness (a name and descriptive paragraph is sufficient). 2. I have learned a uniform random number generator [e.g. not the algorithms: i)Wichmann-Hill, ii) Marsaglia-Multicarry, iii) Super-Duper (Marsaglia), iv) Mersenne-Twister, v) TAOCP-1997 (Knuth), or vi) TAOCP-2002 (Knuth)] . Is there any other method besides that? 3. How to generate 100 random standard normal deviates using the Box-Muller method for standard normal random deviates and sort the sequence, smallest to largest? Your kind help is greatly appreciated. Thanks in advance! best wishes yuying shi [[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] Portable R?
Thanks guys, Dirk: I generally resort to using my laptop for R analysis, but it usually involves dragging loads of data multiple times between a customer system and my laptop. Moving large amounts of data in this fashion can be a problem, particularly when there's sensitivity issues about the data itself. I've used Quantian in the past, and I've customised my own Knoppix CDs, but ideally I'd like to run *my* tools (i.e. R) on *their* system so I'm not having to deal with the logistical and security issues that come with moving data around between systems. I didn't realise that the Windows version of R was relocatable that easily - I just assumed the install did something more complex than that. I'll try copying the R files to a USB key later today. Thanks again Dave M. On 11/17/05, Dirk Eddelbuettel [EMAIL PROTECTED] wrote: On 17 November 2005 at 14:16, David Mitchell wrote: | Hello list, | | A short time ago, I found | http://johnhaller.com/jh/useful_stuff/portable_apps_suite/, which | contains basically a complete set of office tools that can be run | *entirely* from a USB key. The concept is: | - find a Windows PC | - put in your USB key | - run OpenOffice, Firefox, Gaim, Nvu, Thunderbird, ... directly from | your USB key, with no app installation required | - save your files wherever | - remove your USB key and leave, with nothing installed on the original PC | | As a consultant who battles regularly with limited toolsets at | customer sites, this strikes me as an extremely handy way of working. | | Has anyone managed to setup a base R configuration that runs entirely | from USB key? Being a regular user, but no expert, with R, it'd be | very helpful for me if such a mechanism existed, but I've got no idea | where to begin in building such a thing. Short answer: Yes but using Linux, requiring a larger USB stick and some fiddling. Longer answer: Quantian (http://dirk.eddelbuettel.com/quantian) is a everything, the kitchen sink and some Linux distribution running off a DVD. Quantian is focussed on quantitative / numeric apps, and tends to include R plus related goodies -- the last release had an almost complete set of CRAN and BioConductor packages. The raw size of the last release is around 2 GB, corresponding to 6.6 GB expanded. Marco Caliari, who often contributes improved boot code to Quantian, has managed to boot Quantian off a USB stick. I didn't manage to do that with my laptop, possibly because of limitations in its bios. Some of this was discussed in past threads on the quantian-general mailing list. Lots-o-work suggestion: To not require a huge USB stick, you could try to shrink a given live cdrom such as Knoppix or Ubuntu, then add R and other goodies such that you're left with around 512 MB compressed. Then throw it onto a USB stick and make it bootable. Shortcut: Order a Quantian DVD. Some folks sell them pre-made for less than $5. Experiment with that, If you like it, consider making your own mini-distro. Or stick with the DVD and use it directly with the USB stick for your configuration, data, demos, ... Even shorter: R is perfectly relocatable. If you install the Windows binary onto the USB drive, it will run fine. You'll probably need to add editors and other tools. Hope this helps, Dirk -- Statistics: The (futile) attempt to offer certainty about uncertainty. -- Roger Koenker, 'Dictionary of Received Ideas of Statistics' __ R-help@stat.math.ethz.ch 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] Portable R?
I use SLAX on my USB stick (1GB) and I carry R, a host of contributed R packages, all the core BioC packages, rkward as the R frontend, Octave, and many many more. SLAX is in my experience much smaller, lighter, faster, and more complete than any of the other Linux Live CDs out there (my installation with all these packages is about ~600 MB). Heck, I even installed my remastered SLAX onto my iPod nano! Give it a shot. --Jake On Nov 16, 2005, at 9:11 PM, David Mitchell wrote: Thanks guys, Dirk: I generally resort to using my laptop for R analysis, but it usually involves dragging loads of data multiple times between a customer system and my laptop. Moving large amounts of data in this fashion can be a problem, particularly when there's sensitivity issues about the data itself. I've used Quantian in the past, and I've customised my own Knoppix CDs, but ideally I'd like to run *my* tools (i.e. R) on *their* system so I'm not having to deal with the logistical and security issues that come with moving data around between systems. I didn't realise that the Windows version of R was relocatable that easily - I just assumed the install did something more complex than that. I'll try copying the R files to a USB key later today. Thanks again Dave M. On 11/17/05, Dirk Eddelbuettel [EMAIL PROTECTED] wrote: On 17 November 2005 at 14:16, David Mitchell wrote: | Hello list, | | A short time ago, I found | http://johnhaller.com/jh/useful_stuff/portable_apps_suite/, which | contains basically a complete set of office tools that can be run | *entirely* from a USB key. The concept is: | - find a Windows PC | - put in your USB key | - run OpenOffice, Firefox, Gaim, Nvu, Thunderbird, ... directly from | your USB key, with no app installation required | - save your files wherever | - remove your USB key and leave, with nothing installed on the original PC | | As a consultant who battles regularly with limited toolsets at | customer sites, this strikes me as an extremely handy way of working. | | Has anyone managed to setup a base R configuration that runs entirely | from USB key? Being a regular user, but no expert, with R, it'd be | very helpful for me if such a mechanism existed, but I've got no idea | where to begin in building such a thing. Short answer: Yes but using Linux, requiring a larger USB stick and some fiddling. Longer answer: Quantian (http://dirk.eddelbuettel.com/quantian) is a everything, the kitchen sink and some Linux distribution running off a DVD. Quantian is focussed on quantitative / numeric apps, and tends to include R plus related goodies -- the last release had an almost complete set of CRAN and BioConductor packages. The raw size of the last release is around 2 GB, corresponding to 6.6 GB expanded. Marco Caliari, who often contributes improved boot code to Quantian, has managed to boot Quantian off a USB stick. I didn't manage to do that with my laptop, possibly because of limitations in its bios. Some of this was discussed in past threads on the quantian-general mailing list. Lots-o-work suggestion: To not require a huge USB stick, you could try to shrink a given live cdrom such as Knoppix or Ubuntu, then add R and other goodies such that you're left with around 512 MB compressed. Then throw it onto a USB stick and make it bootable. Shortcut: Order a Quantian DVD. Some folks sell them pre-made for less than $5. Experiment with that, If you like it, consider making your own mini-distro. Or stick with the DVD and use it directly with the USB stick for your configuration, data, demos, ... Even shorter: R is perfectly relocatable. If you install the Windows binary onto the USB drive, it will run fine. You'll probably need to add editors and other tools. Hope this helps, Dirk -- Statistics: The (futile) attempt to offer certainty about uncertainty. -- Roger Koenker, 'Dictionary of Received Ideas of Statistics' __ R-help@stat.math.ethz.ch 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] Portable R?
On 17 November 2005 at 15:11, David Mitchell wrote: | I didn't realise that the Windows version of R was relocatable that | easily - I just assumed the install did something more complex than | that. I'll try copying the R files to a USB key later today. Yup. I've installed it onto Windows SMB shares to provide little R apps with tcl/tk apps to colleagues. Works very well until you hit snags like requiring ODBC entry on each client which one can't script ... (as far as I know). Cheers, Dirk -- Statistics: The (futile) attempt to offer certainty about uncertainty. -- Roger Koenker, 'Dictionary of Received Ideas of Statistics' __ R-help@stat.math.ethz.ch 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] Morans I for Spatial Surveillance
Hello, I am interested in using Morans I for different time intervals to detect disease clusters. Ultimately I would like to use CUSUM - or similar monitoring statistic to monitor the results of Morans I - similar to the work by Rogerson (2005) Spatial Surveillance and Cummulative Sum Methods in Spatial and Syndromic Surveillance for Public Health. Thus far - thanks to the list I have Morans I running in a loop for each day (however I found that on some days no data is recorded, this caused an error to get around this error I included an elseif statement to skip the calculation for days where no disease notifications are recorded.) I am just wondering if anyone has code/advice for the best way to apprach the next stage of monitoring Morans I over different time intervals for the detection of abnormal clusters. My code thus far is below - I would also like to assemble the results in a neat table if anyone has tips for this I would alo appreciate it as I am rather new to R! Thanks Serryn T1-read.dbf(ob.dbf) N - colnames(T1) T2nb - read.gal(PC.gal,override=TRUE) col.W - nb2listw(T2nb, style=W) for(i in seq(N)) { Vec1 - spNamedVec(N[i+7], T1) Svec1-sum(Vec1) ifelse ( Svec1 0, res[i]-moran.test(spNamedVec(N[i+7], T1), col.W, zero.policy=TRUE), res[i]-NULL) } print(res) [[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] some questions
Firstly, these are all very basic questions, thah you coukld probobly answer yourself by searching the manual and help files. Nevertheless, here are the answers to give you a little head start. Dear R expert, The following is my questions: 1. How to generate two sequences of 150 uniform deviates, called v1 and v2, in the range [-1,1]. v1-runif(n=150,min=-1,max=1) v2-runif(n=150,min=-1,max=1) 2. How to compute r=(v1)^2+(v2)^2 r-(v1)^2+(v2)^2 3.If r is outside the range of (0,1) then it will be discarded. newr-r[r1] newr-newr[r0] 4 How to compute (v1)*sqrt(-2*log(r)/r) and output the first 100 z values in the list. z-(v1)*sqrt(-2*log(r)/r) z[1:100] #or print(z[1:100]) 5. Plot histograms for the sequences from steps 1 and 2. hist(v1) hist(v2) hist(r) Thanks very much for your help! I hope this helps, Ales Ziberna xingyu __ R-help@stat.math.ethz.ch 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] ECDF values
Dear UseRs, maybe is a silly question: how can I get Empirical CDF values from an object created with ecdf()?? Using print I obtain: Empirical CDF Call: ecdf(t) x[1:57] =4.1,4.4,4.5, ..., 491.3, 671.27 Thanks in advance. Regards, Vito Diventare costruttori di soluzioni Became solutions' constructors The business of the statistician is to catalyze the scientific learning process. George E. P. Box Statistical thinking will one day be as necessary for efficient citizenship as the ability to read and write H. G. Wells Top 10 reasons to become a Statistician 1. Deviation is considered normal 2. We feel complete and sufficient 3. We are 'mean' lovers 4. Statisticians do it discretely and continuously 5. We are right 95% of the time 6. We can legally comment on someone's posterior distribution 7. We may not be normal, but we are transformable 8. We never have to say we are certain 9. We are honestly significantly different 10. No one wants our jobs Visitate il portale http://www.modugno.it/ e in particolare la sezione su Palese http://www.modugno.it/archivio/palesesanto_spirito/ __ R-help@stat.math.ethz.ch 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] update R packages in local repos
On Wed, 16 Nov 2005, P Ehlers wrote: This should work: update.packages(ask = graphics, repos = NULL, contriburl = file:///g:/myFolder/myRepository)) Only if the so-called repository contains (only) binary builds under R 2.2.x of packages for Windows. It would be better to set up the 'repository' correctly as a repository. See my article in R-news 5/1 and the R-admin manual for the format of a repository. -peter Muhammad Subianto wrote: I try to update R packages via my local repository. I put all R packages in g:/myFolder/myRepository, I do like Are these source packages or binary packages or what? library(tools) write_PACKAGES(g:/myFolder/myRepository) options(repos=c(LocalR=file://g:/myFolder/myRepository)) getOption(repos) LocalR file://g:/myFolder/myRepository Your syntax is incorrect here: Peter has silently corrected it. file:// syntax has an element for 'machine' followed by a third slash (although 'machine' is not supported in R, so it should be empty). update.packages(ask = graphics) Error in gzfile(file, r) : unable to open connection In addition: Warning message: cannot open compressed file ':/myFolder/myRepository/bin/windows/contrib/2.2/PACKAGES' ?update.packages It produces these file in g:/myFolder/myRepository PACKAGES PACKAGES.gz Could I make this folder (bin/windows/contrib/2.2/)? Why? -- 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] nlme question
On 11/16/05, Wassell, James T., Ph.D. [EMAIL PROTECTED] wrote: I am using the package nlme to fit a simple random effects (variance components model) with 3 parameters: overall mean (fixed effect), between subject variance (random) and within subject variance (random). So to paraphrase, your model can be written as (with the index i representing subject) y_ij = \mu + b_i + e_ij where b_i ~ N(0, \tao^2) e_ij ~ N(0, \sigma_2) and all b_i's and e_ij's are mutually independent. The model has, as you say, 3 parameters, \mu, \tao and \sigma. I have 16 subjects with 1-4 obs per subject. I need a 3x3 variance-covariance matrix that includes all 3 parameters in order to compute the variance of a specific linear combination. Can you specify the 'linear combination' that you want to estimate in terms of the model above? 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] Portable R?
See the rw-FAQ Q2.6 for a more precise answer than has yet appeared in this thread. A while back (for 2.1.0) some aspects of R were optimized for usage from a slow (and possibly read-only) drive. This does presume that the sites you visit will allow you to run programs from an external drive, and in my experience that is often explicitly disallowed. On Thu, 17 Nov 2005, David Mitchell wrote: Hello list, A short time ago, I found http://johnhaller.com/jh/useful_stuff/portable_apps_suite/, which contains basically a complete set of office tools that can be run *entirely* from a USB key. The concept is: - find a Windows PC - put in your USB key - run OpenOffice, Firefox, Gaim, Nvu, Thunderbird, ... directly from your USB key, with no app installation required - save your files wherever - remove your USB key and leave, with nothing installed on the original PC As a consultant who battles regularly with limited toolsets at customer sites, this strikes me as an extremely handy way of working. Has anyone managed to setup a base R configuration that runs entirely from USB key? Being a regular user, but no expert, with R, it'd be very helpful for me if such a mechanism existed, but I've got no idea where to begin in building such a thing. Thanks in advance for any responses or suggestions -- 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] generalised linear mixed effect model, glmmPQL
Dear listers, I am trying to get more familiar with concepts underlying generalised linear mixed models, mainly through Venables and Ripley (fourth edition) and the R-list archive. Of course, as a possibly tool-user biologist I am not that easy with every détails of the mathematical aspects of the optimisation methods described. I am trying to sort out which could be an acceptable strategy for model comparisons and selection. I have understood from the R-list archive that AIC and similar approaches are not valid for model comparisons with PQL. On the other hand, table 10.4 in Venables and Ripley compares the results of various fitting methods (thus the fitting methods), but my wonder is about comparing models (or parameter estimates) given a method. Can somebody put me on the track with some hints or basic references for dummies if any... Thanks in advance, Patrick __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html