Re: [R] Importing IDL Structures
Michael Lefsky [EMAIL PROTECTED] writes: I am trying to get started with R, but before I do, I need to find a way to import my existing datasets, which are currently stored as arrays of IDL structures (RSI's IDL, not the other one). The problem I has is this: the IDL structures contain scalar items, as well as n-dimensional arrays. I can export the data in a number of ways, including as separate files for scalars and for each of the arrays. Has anyone tackled this problem? If not, can you advise me on the best data structure(s) to hold such data in R? Data frames seemed to be the most obvious choice, but I prefer the syntax used for lists (it is more similar to IDL), if that is possible. And of course (as I said in a previous message) there is the question of how to import the data into the structure (it looks like I will need to export each array separately , import them into R and then assemble the final structure). I don't know IDL, but it seems to do HDF. Would the hdf5 package help, I wonder? -- 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] Graphical presentation of logistic regression
On Wed, 2005-09-14 at 06:29 -0500, Frank E Harrell Jr wrote: Beale, Colin wrote: Hi, I wonder if anyone has written any code to implement the suggestions of Smart et al (2004) in the Bulletin of the Ecological Society of America for a new way of graphically presenting the results of logistic regression (see www.esapubs.org/bulletin/backissues/085-3/bulletinjuly2004_2column.htm#t ools1 for the full text)? I couldn't find anything relating to this sort of graphical representation of logistic models in the archives, but maybe someone has solved it already? In short, Smart et al suggest that a logistic regression be presented as a combination of the two histograms for successes and failures (with one presented upside down at the top of the figure, the other the right way up at the bottom) overlaid by the probability function (ie logistic curve). It's somewhat hard to describe, but is nicely illustrated in the full text version above. I think it is a sensible way of presenting these results and am keen to do so - at the moment I can only do this by generating the two histograms and the logistic curve separately (using hist() and lines()), then copying and pasting the graphs out of R and inverting one in a graphics package, before overlying the others. I'm sure this could be done within R and would be a handy plotting function to develop. Has anyone done so, or can anyone give me any pointers to doing this? I really nead to know how to invert a histogram and how to overlay this with another histogram the right way up. Any thoughts would be welcome. Thanks in advance, Colin From what you describe, that is a poor way to represent the model except for judging discrimination ability (if the model is calibrated well). Effect plots, odds ratio charts, and nomograms are better. See the Design package for details. You're correct when you say that this is a poor way to represent the model. However, you should have some understanding to us ecologists who are simple creatures working with tangible subjects such as animals and plants (microbiologists work with less tangible things). Therefore we want to have a concrete and simple representation. After all, the example was about occurrence of an animal against a concrete environmental variable, and a concrete representation was suggested. Nomograms and things are abstractions that you understand first after long education and training (I tried the Design package and I didn't understand the nomogram plot). I tried with one concrete example with my own data, and the inverted histogram method was patently misleading (with Baz Rowlingson's neat and compact code, sorry for the repetition). The method would be useful with dense and regular data only, but now the clearest visual cue was the uneven sampling intensity. With my limited knowledge on R facilities, I can now remember only two ways two preserve the concreteness of display in the base R: jitter() to avoid overplotting of observations, and sunflowerplot() to show the amount of overplotting. I think Ecological Society of America would be happy to receive papers to suggest better ways to represent binary response data, if some of the knowledgeable persons in this groups would decided to educate them (I'm not an ESA member, so I wouldn't be educated: therefore 'them' instead of 'us'). The ESA bulletin will be influential in manuscript submitted to the Society journals in the future, and the time for action is now. cheers, jari oksanen -- Jari Oksanen -- Dept Biology, Univ Oulu, 90014 Oulu, Finland Ph. +358 8 5531526, cell +358 40 5136529, fax +358 8 5531061 email [EMAIL PROTECTED], homepage http://cc.oulu.fi/~jarioksa/ __ R-help@stat.math.ethz.ch 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] Long lines with Sweave
Jan == Jan T Kim [EMAIL PROTECTED] on Wed, 14 Sep 2005 14:46:20 +0100 writes: Jan On Wed, Sep 14, 2005 at 02:49:56PM +0200, Henrik Andersson wrote: Jan T. Kim wrote: On Wed, Sep 14, 2005 at 10:14:59AM +0200, Henrik Andersson wrote: I have used Sweave a lot the latest year, but never really used any long function calls. If I have code which look like this - gof - benthic.flux(ID=Gulf of Finland, meas.conc=conc, bw.conc=bw.conc, time=times, substance=expression(DIC~(mmol~m^{-3})) ) - I get the output by Sweave in my pdf file, like this: --- gof - benthic.flux(ID = Gulf of Finland, meas.conc = conc, + bw.conc = bw.conc, time = times, substance = expression(DIC ~ + (mmol ~ m^{ + -3 + }))) I can understand that it will not look exactly as entered but why is the '-3' on a line of it's own? Can anyone suggest a idea to how I can make this more readable. It seems you've been thinking LaTeX rather than R ;-) : The exponent -3 in the expression should be enclosed by parentheses rather than by curly braces. The code formatting done by the print method inserts the newline after { and before }. Best regards, Jan If you look at demo(plotmath), I get the impression that m^(-3) does not give me the desired behavior. I want to have -3 in superscript without visible parentheses. Tricky! Jan Ok, I see. Jan It seems to me that you could omit the curly braces in the example, I Jan don't see any differences between the title in the plots produced by Jan plot(1:10, main = expression(DIC~(mmol~m^-3))) Jan and Jan plot(1:10, main = expression(DIC~(mmol~m^{-3}))) Jan For more complex exponents, you could try plain() to prevent them from Jan being wrongly grouped by operator precedence, as in Jan plot(1:10, main = expression(DIC~(mmol~m^plain(-3 + t neat idea, but Jan Not exactly ideal for readability, however... indeed. And really only a workaround: You shouldn't have to uglify your R code in order to work around Sweave pecularities. Martin __ R-help@stat.math.ethz.ch 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] Scan and Lists
Michael == Michael Lefsky [EMAIL PROTECTED] on Wed, 14 Sep 2005 15:06:17 -0600 writes: Michael This may be a newbie question - although I did Michael search for this error message in the archives and Michael via google and didn't see this error: I know how useful google can be - - still, sometimes one would better spend the time differently ... You know the old IBM motto? If not, google for IBM motto ;-) Michael The help page for scan indicates that among the types of data Michael capable of being read are: The supported types are 'logical', 'integer', 'numeric', 'complex', 'character', 'raw' and 'list': 'list' values should have elements which are one of the first six types listed or 'NULL'. Michael I have tried to use a list within a what list : which was wrong: Michael f - scan(file=c:/test/testout.csv, Michael what=list(hi=0.0,bye=,wave=list(1:1000)), Michael sep=,,skip=1) Michael and the following error is returned: Michael c:/test/testout.csv, what = list(hi = 0, bye = , : Michael unimplemented type 'list' in 'extractItem' Michael So, is my syntax confusing R, or is the documentation wrong, or is it Michael some other, third, option? 3rd: You didn't read the documentation carefully enough (though I agree that the current wording leaves a non-small possibility for confusion): In your above citation from the help page, you've left off crucial context. Here is a bit more what: the type of 'what' gives the type of data to be read. If 'what' is a list, it is assumed that the lines of the data file are records each containing 'length(what)' items (fields). The supported types are 'logical', 'integer', 'numeric', 'complex', 'character', 'raw' and 'list': 'list' values should have elements which are one of the first six types listed or 'NULL'. So 'what' has a *type* and that type can be logical, , and list. where list should have elements from the first six types --- which do *NOT* include list. In short: It does exclude using a *list* inside the list. Is your data double character int int int with 1000 integers, i.e. you have 1002 columns? If yes, you'd probably get what you want by whatCols - c(list(hi=0.0, bye=), as.list(1:1000)) f - scan(file = c:/test/testout.csv, what = whatCols, sep= ,, skip=1) {The point here is that c(l1, l2) is used to concatenate two lists l1 and l2; and yes: Please do use spaces {and indentation} to make your more readable ! } Michael Thanks You're welcome, 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] if() command
__ R-help@stat.math.ethz.ch 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: deleting rows
hi all hopefully some one can help. assume that i imported the following data into R (say the data frame is called a) x1 x2 x3 1 NA 3 1 2 NA 1 2 3 3 NA 6 4 5 9 7 5 6 7 8 9 NA 7 9 How can i construct a new data frame that only contains those rows that does not contain the NA's? is these a quick way? ie x1 x2 x3 1 2 3 4 5 9 7 5 6 7 8 9 in this example we can simple use a[c(-1,-2,-4,-8),] but happens if we have a larger dataframe? we need to construct some kind of row indicator telling R which rows contains NA'S. is there an easier method? / allan__ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R: deleting rows
look at function ?complete.cases(), e.g., a[complete.cases(a), ] will do the work in your case. 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: +32/16/336899 Fax: +32/16/337015 Web: http://www.med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: Clark Allan [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Cc: Birgit Erni [EMAIL PROTECTED] Sent: Thursday, September 15, 2005 10:22 AM Subject: [R] R: deleting rows hi all hopefully some one can help. assume that i imported the following data into R (say the data frame is called a) x1 x2 x3 1 NA 3 1 2 NA 1 2 3 3 NA 6 4 5 9 7 5 6 7 8 9 NA 7 9 How can i construct a new data frame that only contains those rows that does not contain the NA's? is these a quick way? ie x1 x2 x3 1 2 3 4 5 9 7 5 6 7 8 9 in this example we can simple use a[c(-1,-2,-4,-8),] but happens if we have a larger dataframe? we need to construct some kind of row indicator telling R which rows contains NA'S. is there an easier method? / allan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R: deleting rows
Clark Allan wrote: hi all hopefully some one can help. assume that i imported the following data into R (say the data frame is called a) x1x2 x3 1 NA 3 1 2 NA 1 2 3 3 NA 6 4 5 9 7 5 6 7 8 9 NA7 9 How can i construct a new data frame that only contains those rows that does not contain the NA's? is these a quick way? ie x1x2 x3 1 2 3 4 5 9 7 5 6 7 8 9 in this example we can simple use a[c(-1,-2,-4,-8),] but happens if we have a larger dataframe? we need to construct some kind of row indicator telling R which rows contains NA'S. is there an easier method? na.omit(a) Uwe Ligges / allan __ R-help@stat.math.ethz.ch 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] Long lines with Sweave
On Thu, 15 Sep 2005 09:35:25 +0200, Martin Maechler (MM) wrote: Jan == Jan T Kim [EMAIL PROTECTED] on Wed, 14 Sep 2005 14:46:20 +0100 writes: Jan On Wed, Sep 14, 2005 at 02:49:56PM +0200, Henrik Andersson wrote: Jan T. Kim wrote: On Wed, Sep 14, 2005 at 10:14:59AM +0200, Henrik Andersson wrote: I have used Sweave a lot the latest year, but never really used any long function calls. If I have code which look like this - gof - benthic.flux(ID=Gulf of Finland, meas.conc=conc, bw.conc=bw.conc, time=times, substance=expression(DIC~(mmol~m^{-3})) ) - I get the output by Sweave in my pdf file, like this: --- gof - benthic.flux(ID = Gulf of Finland, meas.conc = conc, + bw.conc = bw.conc, time = times, substance = expression(DIC ~ + (mmol ~ m^{ + -3 + }))) I can understand that it will not look exactly as entered but why is the '-3' on a line of it's own? Can anyone suggest a idea to how I can make this more readable. It seems you've been thinking LaTeX rather than R ;-) : The exponent -3 in the expression should be enclosed by parentheses rather than by curly braces. The code formatting done by the print method inserts the newline after { and before }. Best regards, Jan If you look at demo(plotmath), I get the impression that m^(-3) does not give me the desired behavior. I want to have -3 in superscript without visible parentheses. Tricky! Jan Ok, I see. Jan It seems to me that you could omit the curly braces in the example, I Jan don't see any differences between the title in the plots produced by Jan plot(1:10, main = expression(DIC~(mmol~m^-3))) Jan and Jan plot(1:10, main = expression(DIC~(mmol~m^{-3}))) Jan For more complex exponents, you could try plain() to prevent them from Jan being wrongly grouped by operator precedence, as in Jan plot(1:10, main = expression(DIC~(mmol~m^plain(-3 + t neat idea, but Jan Not exactly ideal for readability, however... indeed. And really only a workaround: You shouldn't have to uglify your R code in order to work around Sweave pecularities. Hmm, it's not really an Sweave peculiarity, but one of the R parser. After saving Henrik's code in file test.R I get R x = parse(test.R) R x expression(gof - benthic.flux(ID = Gulf of Finland, meas.conc = conc, bw.conc = bw.conc, time = times, substance = expression(DIC ~ (mmol ~ m^{ -3 } R deparse(x) [1] expression(gof - benthic.flux(ID = \Gulf of Finland\, meas.conc = conc, [2] bw.conc = bw.conc, time = times, substance = expression(DIC ~ [3] (mmol ~ m^{ [4] -3 [5] } and the latter is used by Sweave. The code chunks need to be parsed, because otherwise there is no way how to know where to insert output. A source(..., echo=TRUE) will suffer from the same problem. Best, Fritz -- --- Friedrich Leisch Institut für Statistik Tel: (+43 1) 58801 10715 Technische Universität WienFax: (+43 1) 58801 10798 Wiedner Hauptstraße 8-10/1071 A-1040 Wien, Austria http://www.ci.tuwien.ac.at/~leisch __ R-help@stat.math.ethz.ch 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] Long lines with Sweave
Fritz == Friedrich Leisch [EMAIL PROTECTED] on Thu, 15 Sep 2005 10:36:43 +0200 writes: On Thu, 15 Sep 2005 09:35:25 +0200, Martin Maechler (MM) wrote: Jan == Jan T Kim [EMAIL PROTECTED] on Wed, 14 Sep 2005 14:46:20 +0100 writes: Jan On Wed, Sep 14, 2005 at 02:49:56PM +0200, Henrik Andersson wrote: Jan T. Kim wrote: On Wed, Sep 14, 2005 at 10:14:59AM +0200, Henrik Andersson wrote: I have used Sweave a lot the latest year, but never really used any long function calls. If I have code which look like this - gof - benthic.flux(ID=Gulf of Finland, meas.conc=conc, bw.conc=bw.conc, time=times, substance=expression(DIC~(mmol~m^{-3})) ) - I get the output by Sweave in my pdf file, like this: --- gof - benthic.flux(ID = Gulf of Finland, meas.conc = conc, + bw.conc = bw.conc, time = times, substance = expression(DIC ~ + (mmol ~ m^{ + -3 + }))) I can understand that it will not look exactly as entered but why is the '-3' on a line of it's own? Can anyone suggest a idea to how I can make this more readable. It seems you've been thinking LaTeX rather than R ;-) : The exponent -3 in the expression should be enclosed by parentheses rather than by curly braces. The code formatting done by the print method inserts the newline after { and before }. Best regards, Jan If you look at demo(plotmath), I get the impression that m^(-3) does not give me the desired behavior. I want to have -3 in superscript without visible parentheses. Tricky! Jan Ok, I see. Jan It seems to me that you could omit the curly braces in the example, I Jan don't see any differences between the title in the plots produced by Jan plot(1:10, main = expression(DIC~(mmol~m^-3))) Jan and Jan plot(1:10, main = expression(DIC~(mmol~m^{-3}))) Jan For more complex exponents, you could try plain() to prevent them from Jan being wrongly grouped by operator precedence, as in Jan plot(1:10, main = expression(DIC~(mmol~m^plain(-3 + t neat idea, but Jan Not exactly ideal for readability, however... indeed. And really only a workaround: You shouldn't have to uglify your R code in order to work around Sweave pecularities. Fritz Hmm, it's not really an Sweave peculiarity, but one of the R Fritz parser. Indeed, of course -- it's R's internal deparse(.) : Here's a smaller example: (cc - expression(x ^ {-3})[[1]]) str(as.list(cc)) List of 3 $ : symbol ^ $ : symbol x $ : language { -3 } Please apologize for any bad light that I might have shed on Sweave. I do love its concept! Martin __ R-help@stat.math.ethz.ch 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 sort data sets?
Hi, I was wondering if someone know how to sort a data set by column. I've tried sort() but without luck. I would think there should be a function for it somewhere. An example with the iris data set would be appreciated. Thanks, Martin __ R-help@stat.math.ethz.ch 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] How to sort data sets?
see : order, sort.list, sort and rank hih __ R-help@stat.math.ethz.ch 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] How to sort data sets?
Martin Lam wrote: Hi, I was wondering if someone know how to sort a data set by column. I've tried sort() but without luck. I would think there should be a function for it somewhere. An example with the iris data set would be appreciated. See ?order. Uwe Ligges Thanks, Martin __ R-help@stat.math.ethz.ch 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] Rcommander and simple chisquare
In this years biostat teaching I will include Rcommander (it indeed simplifies syntax problems that makes students frequently miss the core statistical problems). But I could not find how to make a simple chisquare comparison between observed frequencies and expected frequencies (eg in genetics where you expect phenotypic frequencies corresponding to 3:1 in standard dominant/recessif alleles). Any idea where this feature might be hidden? Or could it be added to Rcommander? Thanks, Christian. ps: in case I am not making myself clear, can Rcommander be made to perform chisq.test(c(61,39),p=c(0.75,0.25)) __ R-help@stat.math.ethz.ch 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] means comparison in R (post-hoc test)
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi. I have been using SAS for some time, and now I have discovered R. I am very happy with it, but I have not found out how to perform some of the multiple comparisons I was used to do in SAS. With the SAS/STAT, I generally used the MEANS (for comparison of arithmetic means) and the LSMEANS (for adjusted means) statements of the GLM procedure (I think it is equivalent to lm in R). They provided a lot of tests: LSD, Duncan, Tukey-Kramer, Bonferroni, Scheffé, SNK, etc. However, in R I have only discovered Tukey-HSD. I have searched for information about this, but I was not successful. I wonder if anybody knows where I could learn about this. I would like to use these tests in R, and also obtain the adjusted means like these produced by the LSMEANS statement. Thank you. Felipe - --oOo-- Felipe Martínez-Pastor (BSc, PhD) Animal Reproduction and Obstetrics Veterinary Clinic Hospital 24071-León (Spain) Phone: 987 291 430 / 987 291 000 + 5203 Fax: 987 295 203 e-mail: [EMAIL PROTECTED] - --- This message has been electronically signed using GPG, a free system to sign and encrypt documents. If you want to learn more, visit http://www.gnupg.org -BEGIN PGP SIGNATURE- iEYEARECAAYFAkMpUVYACgkQWtdQtNzjBl4rywCePiSxJw8/N6HzdZ7C+YMHf2K6 YSYAniw5GSo0ihrt4+OabHJ4c2PKNHkp =aF38 -END PGP SIGNATURE- __ R-help@stat.math.ethz.ch 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] if() command
Hi Sorry, I am not sure why sometimes is a text from my answeres stripped off. On 14 Sep 2005 at 14:09, Carlos Mauricio Cardeal Mende wrote: Ok Petr, I run your suggestion and I got this message: age-sample(seq(10,50,10), 20, replace=T) if (age =10) {group - 1} else if (age 10 age = 20) {group - 2} else {group - 3} Warning message: the condition has length 1 and only the first element will be used in: if (age = 10) { What does it means ? Others has already answered it but I told you that if command is not vhat you probably want and suggested to use cut e.g. cutvector - c(0,10,20,100) group - as.numeric(cut(age,cutvector)) You can change your cutvector according to your wish anytime before calling the second function HTH Petr And when I look to the database I have no new classification ! Could you help please ? Mauricio Petr Pikal escreveu: Hallo On 13 Sep 2005 at 10:29, Carlos Maurício Cardeal Mende wrote: Hi everyone ! Could you please help me with this problem ? I´ve trying to write a code that assign to a variable the content from another, but all I´ve got is a message error. For example: if (age =10) {group == 1} else if (age 10 age = 20) {group == 2} else {group == 3} if you put your statement on one line it works (at least it does not give you syntax error) but the result is hardly what you really expect age-sample(seq(10,50,10), 20, replace=T) if (age =10) {group - 1} else if (age 10 age = 20) {group - 2} else {group - 3} if (age =10) {group == 1} else if (age 10 age = 20) {group == 2} else {group == 3} Maybe you want something like group-as.numeric(cut(age,c(0,10,20,100))) but it is only guess HTH Petr Syntax error Or if (age =10) {group == 1} else (age 10 age = 20) {group == 2} else {group == 3} Syntax error I know that is possible to find the solution by ifelse command or even recode command, but I´d like to use this way, because I can add another variable as a new condition and I believe to expand the possibilites. Thanks, Mauricio __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Petr Pikal [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] stripped postings when not following the posting guide
Petr == Petr Pikal [EMAIL PROTECTED] on Thu, 15 Sep 2005 13:27:43 +0200 writes: Petr Hi Petr Sorry, I am not sure why sometimes is a text from my answeres Petr stripped off. Hi Petr, it's when you don't follow the posting guide _and_ simultaneously happen to fool the filters so that something is posted at all: The one you mention from this morning was full of HTML crap, (wc on the body gave 302 1491 17036), but wasn't correctly recognized as HTML by one part of the filters but by the others (that strip HTML). [ If the filters worked as intended currently, such a message would be completely swallowed by the filters! ] Regards, Martin __ R-help@stat.math.ethz.ch 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] Rcommander and simple chisquare
Hello, Just look at Statistics - Contingency tables. There is an option for making the chi square test there. Best, Philippe Grosjean, ..°})) ) ) ) ) ) ( ( ( ( (Prof. Philippe Grosjean ) ) ) ) ) ( ( ( ( (Numerical Ecology of Aquatic Systems ) ) ) ) ) Mons-Hainaut University, Pentagone (3D08) ( ( ( ( (Academie Universitaire Wallonie-Bruxelles ) ) ) ) ) 8, av du Champ de Mars, 7000 Mons, Belgium ( ( ( ( ( ) ) ) ) ) phone: + 32.65.37.34.97, fax: + 32.65.37.30.54 ( ( ( ( (email: [EMAIL PROTECTED] ) ) ) ) ) ( ( ( ( (web: http://www.umh.ac.be/~econum ) ) ) ) ) http://www.sciviews.org ( ( ( ( ( .. Christian Jost wrote: In this years biostat teaching I will include Rcommander (it indeed simplifies syntax problems that makes students frequently miss the core statistical problems). But I could not find how to make a simple chisquare comparison between observed frequencies and expected frequencies (eg in genetics where you expect phenotypic frequencies corresponding to 3:1 in standard dominant/recessif alleles). Any idea where this feature might be hidden? Or could it be added to Rcommander? Thanks, Christian. ps: in case I am not making myself clear, can Rcommander be made to perform chisq.test(c(61,39),p=c(0.75,0.25)) __ R-help@stat.math.ethz.ch 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] Graphical presentation of logistic regression
If a graphical presentation provides improved insight then that is sufficient justification. The existence of better more precise methods, does not change that. I, too, sometimes use jitter() to avoid overplotting of observations, but I think the dot-plots in de la Cruz's code are even better. It is the histogram that is misleading (due to paucity of data), not the effort to elucidate the joint behavior of zeros and ones. http://www.esapubs.org/bulletin/backissues/086-1/bulletinjan2005.htm#et Please try a variation that his code provides: plot.logi.hist(independ = altitude, depend = tree, logi.mod = 1, type = dit, boxp = TRUE, rug = TRUE, las.h = 1) which does not use the histograms but instead uses dit plots to provide a helpful, visceral feel for the behavior of the observations. Charles Annis, P.E. [EMAIL PROTECTED] phone: 561-352-9699 eFax: 614-455-3265 http://www.StatisticalEngineering.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jari Oksanen Sent: Thursday, September 15, 2005 3:17 AM To: Frank E Harrell Jr Cc: r-help@stat.math.ethz.ch; Beale, Colin Subject: Re: [R] Graphical presentation of logistic regression On Wed, 2005-09-14 at 06:29 -0500, Frank E Harrell Jr wrote: Beale, Colin wrote: Hi, I wonder if anyone has written any code to implement the suggestions of Smart et al (2004) in the Bulletin of the Ecological Society of America for a new way of graphically presenting the results of logistic regression (see www.esapubs.org/bulletin/backissues/085-3/bulletinjuly2004_2column.htm#t ools1 for the full text)? I couldn't find anything relating to this sort of graphical representation of logistic models in the archives, but maybe someone has solved it already? In short, Smart et al suggest that a logistic regression be presented as a combination of the two histograms for successes and failures (with one presented upside down at the top of the figure, the other the right way up at the bottom) overlaid by the probability function (ie logistic curve). It's somewhat hard to describe, but is nicely illustrated in the full text version above. I think it is a sensible way of presenting these results and am keen to do so - at the moment I can only do this by generating the two histograms and the logistic curve separately (using hist() and lines()), then copying and pasting the graphs out of R and inverting one in a graphics package, before overlying the others. I'm sure this could be done within R and would be a handy plotting function to develop. Has anyone done so, or can anyone give me any pointers to doing this? I really nead to know how to invert a histogram and how to overlay this with another histogram the right way up. Any thoughts would be welcome. Thanks in advance, Colin From what you describe, that is a poor way to represent the model except for judging discrimination ability (if the model is calibrated well). Effect plots, odds ratio charts, and nomograms are better. See the Design package for details. You're correct when you say that this is a poor way to represent the model. However, you should have some understanding to us ecologists who are simple creatures working with tangible subjects such as animals and plants (microbiologists work with less tangible things). Therefore we want to have a concrete and simple representation. After all, the example was about occurrence of an animal against a concrete environmental variable, and a concrete representation was suggested. Nomograms and things are abstractions that you understand first after long education and training (I tried the Design package and I didn't understand the nomogram plot). I tried with one concrete example with my own data, and the inverted histogram method was patently misleading (with Baz Rowlingson's neat and compact code, sorry for the repetition). The method would be useful with dense and regular data only, but now the clearest visual cue was the uneven sampling intensity. With my limited knowledge on R facilities, I can now remember only two ways two preserve the concreteness of display in the base R: jitter() to avoid overplotting of observations, and sunflowerplot() to show the amount of overplotting. I think Ecological Society of America would be happy to receive papers to suggest better ways to represent binary response data, if some of the knowledgeable persons in this groups would decided to educate them (I'm not an ESA member, so I wouldn't be educated: therefore 'them' instead of 'us'). The ESA bulletin will be influential in manuscript submitted to the Society journals in the future, and the time for action is now. cheers, jari oksanen -- Jari Oksanen -- Dept Biology, Univ Oulu, 90014 Oulu, Finland Ph. +358 8 5531526, cell +358 40 5136529, fax +358 8 5531061 email [EMAIL PROTECTED], homepage http://cc.oulu.fi/~jarioksa/ __
[R] how to do sthg like mat[!=(ind),]
Hi I want to do something which seems straightforward, but I couldn't find the way to do this. I have a matrix called m for example, and a vector of values (let's call ind this vector) which are indices of lines I don't want to keep. for example I have: m v1 v2 v3 [1,] 1 4 7 [2,] 2 5 8 [3,] 3 6 9 [4,] 10 11 12 ind [1] 2 4 I would like to obtain this: m2 v1 v2 v3 [1,] 1 4 7 [3,] 3 6 9 by saying something like m2-m[!=(ind),] but this line does not work. Any idea or suggestion highly welcome ! Florence. [[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] stripped postings when not following the posting guide
Hi Martin It sometimes happens when I respond to somebodys question. When I post my own I try to follow the posting guide closely and do not use HTML crap at all. I just do not know how to recognize that my response has this unwanted residues and how to get rid of them before I actually find that the message is stripped off. I am using Pegassus Mail 4.21c. Thanks Petr On 15 Sep 2005 at 14:30, Martin Maechler wrote: Petr == Petr Pikal [EMAIL PROTECTED] on Thu, 15 Sep 2005 13:27:43 +0200 writes: Petr Hi Petr Sorry, I am not sure why sometimes is a text from my answeres Petr stripped off. Hi Petr, it's when you don't follow the posting guide _and_ simultaneously happen to fool the filters so that something is posted at all: The one you mention from this morning was full of HTML crap, (wc on the body gave 302 1491 17036), but wasn't correctly recognized as HTML by one part of the filters but by the others (that strip HTML). [ If the filters worked as intended currently, such a message would be completely swallowed by the filters! ] Regards, Martin __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Petr Pikal [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] how to do sthg like mat[!=(ind),]
Florence Combes wrote: Hi I want to do something which seems straightforward, but I couldn't find the way to do this. I have a matrix called m for example, and a vector of values (let's call ind this vector) which are indices of lines I don't want to keep. for example I have: m v1 v2 v3 [1,] 1 4 7 [2,] 2 5 8 [3,] 3 6 9 [4,] 10 11 12 ind [1] 2 4 I would like to obtain this: m2 v1 v2 v3 [1,] 1 4 7 [3,] 3 6 9 by saying something like m2-m[!=(ind),] See help([) and learn about negative indices: m2 - m[-ind,] Uwe Ligges but this line does not work. Any idea or suggestion highly welcome ! Florence. [[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] means comparison in R (post-hoc test)
Felipe felipe at unileon.es writes: With the SAS/STAT, I generally used the MEANS (for comparison of arithmetic means) and the LSMEANS (for adjusted means) statements of the GLM procedure (I think it is equivalent to lm in R). They provided a lot of tests: LSD, Duncan, Tukey-Kramer, Bonferroni, Scheffé, SNK, etc. However, in R I have only discovered Tukey-HSD. Package multcomp with the workhorse-function simint comes close to what you want. Dieter __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Splitting the string at the last sub-string
Hi, I need to split a string into 2 strings, with the split point defined by the last occurrence of some substring. I come up with some convoluted code to do so: str = Chance favors the prepared mind sub = e y = unlist(strsplit(str,sub)) z = cbind(paste(y[-length(y)], sub, sep=, collapse = ), y[length(y)]); y z z[1] z[2] Is there a simpler way to do so? I think ~8 function calls to do such a simple operation is an overkill. Jarek \ Jarek Tuszynski, PhD. o / \ Science Applications International Corporation \__,| (703) 676-4192 \ [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] Graphical presentation of logistic regression
Jari Oksanen wrote: On Wed, 2005-09-14 at 06:29 -0500, Frank E Harrell Jr wrote: Beale, Colin wrote: Hi, I wonder if anyone has written any code to implement the suggestions of Smart et al (2004) in the Bulletin of the Ecological Society of America for a new way of graphically presenting the results of logistic regression (see www.esapubs.org/bulletin/backissues/085-3/bulletinjuly2004_2column.htm#t ools1 for the full text)? I couldn't find anything relating to this sort of graphical representation of logistic models in the archives, but maybe someone has solved it already? In short, Smart et al suggest that a logistic regression be presented as a combination of the two histograms for successes and failures (with one presented upside down at the top of the figure, the other the right way up at the bottom) overlaid by the probability function (ie logistic curve). It's somewhat hard to describe, but is nicely illustrated in the full text version above. I think it is a sensible way of presenting these results and am keen to do so - at the moment I can only do this by generating the two histograms and the logistic curve separately (using hist() and lines()), then copying and pasting the graphs out of R and inverting one in a graphics package, before overlying the others. I'm sure this could be done within R and would be a handy plotting function to develop. Has anyone done so, or can anyone give me any pointers to doing this? I really nead to know how to invert a histogram and how to overlay this with another histogram the right way up. Any thoughts would be welcome. Thanks in advance, Colin From what you describe, that is a poor way to represent the model except for judging discrimination ability (if the model is calibrated well). Effect plots, odds ratio charts, and nomograms are better. See the Design package for details. You're correct when you say that this is a poor way to represent the model. However, you should have some understanding to us ecologists who are simple creatures working with tangible subjects such as animals and plants (microbiologists work with less tangible things). Therefore we want to have a concrete and simple representation. After all, the example was about occurrence of an animal against a concrete environmental variable, and a concrete representation was suggested. Nomograms and things are abstractions that you understand first after long education and training (I tried the Design package and I didn't understand the nomogram plot). I don't understand why you think the histograms are representing the model. That approach even seems to be interchanging the roles of the independent and dependent variables. I tried with one concrete example with my own data, and the inverted histogram method was patently misleading (with Baz Rowlingson's neat and compact code, sorry for the repetition). The method would be useful with dense and regular data only, but now the clearest visual cue was the uneven sampling intensity. With my limited knowledge on R facilities, I can now remember only two ways two preserve the concreteness of display in the base R: jitter() to avoid overplotting of observations, and sunflowerplot() to show the amount of overplotting. I think Ecological Society of America would be happy to receive papers to suggest better ways to represent binary response data, if some of the knowledgeable persons in this groups would decided to educate them (I'm not an ESA member, so I wouldn't be educated: therefore 'them' instead of 'us'). The ESA bulletin will be influential in manuscript submitted to the Society journals in the future, and the time for action is now. See @Article{gui00ord, author = {Guisan, Antoine and Harrell, Frank E.}, title = {Ordinal response regression models in ecology}, journal = {Journal of Vegetation Science}, year =2000, volume = 11, pages = {617-626}, annote = {teaching;ordinal logistic model} } This is more complex than needed (ordinal instead of binary) but binary is a special case of ordinal. Cheers, Frank cheers, jari oksanen -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] make check FAILS - Error code 1...comparing d-p-q-r-tests.Rout
Under FreeBSD 5.3, attempting to properly install R-2.1.1, I get the following response when I. % make ;all finishes without error, then... % make check ret . . -- comparing d-p-q-r-tests.Rout to d-p-q-r-tests.Rout.save 1004c1004 [1] mean relative difference 1.2848649e-08 [1] TRUE .Error code 1 stop in ~R/R-2.1.1/tests --- I assume a computed value is out of bounds regarding a predetermined range of accuracy. Not being a statistician nor programmer, how might I fix this, please ? Appreciatively, Courtney __ R-help@stat.math.ethz.ch 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] Splitting the string at the last sub-string
Prof Brian Ripley wrote: substring(str, c(1, 26), c(25,length(str))) nchar(str) surely? regexps can be rather slow though. Here's two functions: byRipley = function(str,sub){ lp=attr(regexpr(paste(.*,sub,sep=),str),'match.length') return(substring(str, c(1, lp+1), c(lp,nchar(str } byJarek = function(str,sub){ y = unlist(strsplit(str,sub)) return(cbind(paste(y[-length(y)], sub, sep=, collapse = ), y[length(y)])) } and a quick test: system.time(for(i in 1:10){byJarek(str,sub)}) [1] 15.55 0.10 16.06 0.00 0.00 system.time(for(i in 1:10){byRipley(str,sub)}) [1] 30.28 0.07 31.86 0.00 0.00 Baz __ R-help@stat.math.ethz.ch 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] Remove vector elements from another vector
Hello, I have two vectors of different lengths. Fx a - 1:9; b - c(4, 5). What is the best way to remove the elements in vector b from vector a so that the result would be a vector with elements c(1,2,3,6,7,8,9)? Best regards, Kalle _ Find masser af gode tilbud på MSN Shopping http://shopping.msn.dk/ __ R-help@stat.math.ethz.ch 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] Remove vector elements from another vector
Karsten Luder wrote: Hello, I have two vectors of different lengths. Fx a - 1:9; b - c(4, 5). What is the best way to remove the elements in vector b from vector a so that the result would be a vector with elements c(1,2,3,6,7,8,9)? setdiff(a, b) Uwe Ligges Best regards, Kalle _ Find masser af gode tilbud på MSN Shopping http://shopping.msn.dk/ __ R-help@stat.math.ethz.ch 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] Remove vector elements from another vector
look at function ?setdiff(), e.g., setdiff(a, b) 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: +32/16/336899 Fax: +32/16/337015 Web: http://www.med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: Karsten Luder [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Sent: Tuesday, September 13, 2005 2:35 PM Subject: [SPAM?] [R] Remove vector elements from another vector Hello, I have two vectors of different lengths. Fx a - 1:9; b - c(4, 5). What is the best way to remove the elements in vector b from vector a so that the result would be a vector with elements c(1,2,3,6,7,8,9)? Best regards, Kalle _ Find masser af gode tilbud på MSN Shopping http://shopping.msn.dk/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Remove vector elements from another vector
Karsten Luder [EMAIL PROTECTED] writes: Hello, I have two vectors of different lengths. Fx a - 1:9; b - c(4, 5). What is the best way to remove the elements in vector b from vector a so that the result would be a vector with elements c(1,2,3,6,7,8,9)? I think we had this on the list no more than a week ago... setdiff(a,b) -- 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] Remove vector elements from another vector
Hi On 13 Sep 2005 at 14:35, Karsten Luder wrote: Hello, I have two vectors of different lengths. Fx a - 1:9; b - c(4, 5). What is the best way to remove the elements in vector b from vector a so that the result would be a vector with elements c(1,2,3,6,7,8,9)? which(!a%in%b) [1] 1 2 3 6 7 8 9 a[which(!a%in%b)] [1] 1 2 3 6 7 8 9 a[(!a%in%b)] [1] 1 2 3 6 7 8 9 HTH Petr Best regards, Kalle _ Find masser af gode tilbud pĺ MSN Shopping http://shopping.msn.dk/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Petr Pikal [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] newbie question
Can someone tell me how I create a vector of numbers where the step isn't 1? i.e. x-(0.0,0.5,1.0,1.5) Thanks 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] Splitting the string at the last sub-string
On Thu, 15 Sep 2005, Barry Rowlingson wrote: Prof Brian Ripley wrote: substring(str, c(1, 26), c(25,length(str))) nchar(str) surely? Yes, or anything larger: I actually tested 1. regexps can be rather slow though. Here's two functions: But that's not the way to do this repeatedly for the same pattern. (It is normally compiling regexps that is slow, and regexpr is vectorized.) Not that I would call 300us `slow'. byRipley = function(str,sub){ lp=attr(regexpr(paste(.*,sub,sep=),str),'match.length') return(substring(str, c(1, lp+1), c(lp,nchar(str } byJarek = function(str,sub){ y = unlist(strsplit(str,sub)) return(cbind(paste(y[-length(y)], sub, sep=, collapse = ), y[length(y)])) } and a quick test: system.time(for(i in 1:10){byJarek(str,sub)}) [1] 15.55 0.10 16.06 0.00 0.00 system.time(for(i in 1:10){byRipley(str,sub)}) [1] 30.28 0.07 31.86 0.00 0.00 Baz __ R-help@stat.math.ethz.ch 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
[R] Graphics 'snapshots' in Linux?
Hi, I'm working on a MEPIS (Debian-based Linux) computer, using the emacs/ESS package to do my R work. I've got some plots that I label interactively using the locate function. With the Windows GUI there is an option to take a snapshot of the graphics output, saving it as an image file. Is there a way to do this with emacs/ESS? Thanks, Tyler __ R-help@stat.math.ethz.ch 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] make check FAILS - Error code 1...comparing d-p-q-r-tests.Rout
The problem is a known one and solved in R 2.1.1-patched. On your OS the result is not as accurate as most, but the tolerance set was too tight so the test failure is not something to worry about. Please install R-patched instead, as it has many bug fixes in place. On Thu, 15 Sep 2005, Courtney Thomas wrote: Under FreeBSD 5.3, attempting to properly install R-2.1.1, I get the following response when I. % make;all finishes without error, then... % make check ret . . -- comparing d-p-q-r-tests.Rout to d-p-q-r-tests.Rout.save 1004c1004 [1] mean relative difference 1.2848649e-08 [1] TRUE .Error code 1 stop in ~R/R-2.1.1/tests --- I assume a computed value is out of bounds regarding a predetermined range of accuracy. Not being a statistician nor programmer, how might I fix this, please ? Appreciatively, Courtney __ R-help@stat.math.ethz.ch 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
[R] Combine vector of different length
Hi group, Is there a quick way to cbind vector of different length? I should have checked out the previous postings but somehow I can't access the list from CRAN website. Thanks for your help. Regards, Kelvin __ R-help@stat.math.ethz.ch 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] newbie question
tom wright a écrit : Can someone tell me how I create a vector of numbers where the step isn't 1? i.e. x-(0.0,0.5,1.0,1.5) seq(a, b, 0.5) ?seq hih __ R-help@stat.math.ethz.ch 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] sequences; was: newbie question
tom wright wrote: Can someone tell me how I create a vector of numbers where the step isn't 1? i.e. x-(0.0,0.5,1.0,1.5) Folks, please read the posting guide and basic documentation! Please use a sensible subject. Don't know how many of therelike questions I have read today... See ?seq Uwe Ligges Thanks 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 __ R-help@stat.math.ethz.ch 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] Graphics 'snapshots' in Linux?
Quoting Tyler Smith [EMAIL PROTECTED]: Hi, I'm working on a MEPIS (Debian-based Linux) computer, using the emacs/ESS package to do my R work. I've got some plots that I label interactively using the locate function. With the Windows GUI there is an option to take a snapshot of the graphics output, saving it as an image file. Is there a way to do this with emacs/ESS? In the graphics menu, you could use ksnapshot. -- Jean-Luc __ R-help@stat.math.ethz.ch 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] Splitting the string at the last sub-string
Thanks for suggestions. I suspect the regexpr version will be better than my version, since I use it to find an string towards the end of a large (up to ~30Mb) test/XML file. Thanks again. Jarek \ Jarek Tuszynski, PhD. o / \ Science Applications International Corporation \__,| (703) 676-4192 \ [EMAIL PROTECTED] ` \ -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Prof Brian Ripley Sent: Thursday, September 15, 2005 10:43 AM To: Barry Rowlingson Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Splitting the string at the last sub-string On Thu, 15 Sep 2005, Barry Rowlingson wrote: Prof Brian Ripley wrote: substring(str, c(1, 26), c(25,length(str))) nchar(str) surely? Yes, or anything larger: I actually tested 1. regexps can be rather slow though. Here's two functions: But that's not the way to do this repeatedly for the same pattern. (It is normally compiling regexps that is slow, and regexpr is vectorized.) Not that I would call 300us `slow'. byRipley = function(str,sub){ lp=attr(regexpr(paste(.*,sub,sep=),str),'match.length') return(substring(str, c(1, lp+1), c(lp,nchar(str } byJarek = function(str,sub){ y = unlist(strsplit(str,sub)) return(cbind(paste(y[-length(y)], sub, sep=, collapse = ), y[length(y)])) } and a quick test: system.time(for(i in 1:10){byJarek(str,sub)}) [1] 15.55 0.10 16.06 0.00 0.00 system.time(for(i in 1:10){byRipley(str,sub)}) [1] 30.28 0.07 31.86 0.00 0.00 Baz __ R-help@stat.math.ethz.ch 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 __ R-help@stat.math.ethz.ch 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] about cutree
Hi Everyone, I'm trying to use cutree to get the clusters after hclust. What I used is: mycluster-cutree(cnclust,h=0.5) Now, my problem is, how can I get the actual clusters? Thanks! Best, Baoqiang Cao __ R-help@stat.math.ethz.ch 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] Rcommander and simple chisquare
Dear Christian, From the Rcmdr menus, select Statistics - Summaries - Frequency distributions, and check the Chisquare goodness of fit test box in the resulting dialog. This will bring up a dialog box where you can enter hypothesized probabilities from which expected frequencies will be calculated. Regards, John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Christian Jost Sent: Thursday, September 15, 2005 5:40 AM To: r-help@stat.math.ethz.ch Subject: [R] Rcommander and simple chisquare In this years biostat teaching I will include Rcommander (it indeed simplifies syntax problems that makes students frequently miss the core statistical problems). But I could not find how to make a simple chisquare comparison between observed frequencies and expected frequencies (eg in genetics where you expect phenotypic frequencies corresponding to 3:1 in standard dominant/recessif alleles). Any idea where this feature might be hidden? Or could it be added to Rcommander? Thanks, Christian. ps: in case I am not making myself clear, can Rcommander be made to perform chisq.test(c(61,39),p=c(0.75,0.25)) __ R-help@stat.math.ethz.ch 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] Rcommander and simple chisquare
Dear Christian, From the Rcmdr menus, select Statistics - Summaries - Frequency distributions, and check the Chisquare goodness of fit test box in the resulting dialog. This will bring up a dialog box where you can enter hypothesized probabilities from which expected frequencies will be calculated. Regards, John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Christian Jost Sent: Thursday, September 15, 2005 5:40 AM To: r-help@stat.math.ethz.ch Subject: [R] Rcommander and simple chisquare In this years biostat teaching I will include Rcommander (it indeed simplifies syntax problems that makes students frequently miss the core statistical problems). But I could not find how to make a simple chisquare comparison between observed frequencies and expected frequencies (eg in genetics where you expect phenotypic frequencies corresponding to 3:1 in standard dominant/recessif alleles). Any idea where this feature might be hidden? Or could it be added to Rcommander? Thanks, Christian. ps: in case I am not making myself clear, can Rcommander be made to perform chisq.test(c(61,39),p=c(0.75,0.25)) __ R-help@stat.math.ethz.ch 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] Log scale in histograms
Can't find any information about this, but others must want to do it. In the example below, the second plot has the desired log scale, but the first does not. Any help appreciated. Well, I had to solve this problem myself. I hope that doesn't prove I should have posted it. I did spend a lot of time on it, both before and after posting. Here is a reasonable-looking histogram based on logged data: hist(log10(area_Mh), 12, xlab=Area (Mh), main=, axes=FALSE) axis(1, labels=formatC(10^(axTicks(1)), digits=3)) axis(2) JD __ R-help@stat.math.ethz.ch 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] Graphics 'snapshots' in Linux?
On Thu, 2005-09-15 at 10:43 -0400, Tyler Smith wrote: Hi, I'm working on a MEPIS (Debian-based Linux) computer, using the emacs/ESS package to do my R work. I've got some plots that I label interactively using the locate function. With the Windows GUI there is an option to take a snapshot of the graphics output, saving it as an image file. Is there a way to do this with emacs/ESS? Thanks, Tyler Tyler, Take a look at ?dev.copy2eps and on the same page dev.copy(), which enable you to copy the current X11 plot supported output devices. You could do something like the following for an EPS file: plot(1:5) text(locator(1), Place Text Here) dev.copy2eps(file = MyPlot.eps) or the following for a PNG file: par(bg = white) plot(1:5) text(locator(1), Place Text Here) dev.copy(device = png, file = MyPlot.png) dev.off() Note that in the first example, dev.off() is not required, as the EPS output device is closed after the call. Also, note in the second example, you will need to set the background to white (unless already specified for whatever color you may be using), as the default output file will have a transparent background, even though the png() function shows the default as white. If my memory is correct this is because the X11 device itself has a transparent background by default and this is what is copied. 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
Re: [R] means comparison in R (post-hoc test)
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Thank you, I think multcomp is very near to what I was looking for. However, I am still looking for a mean to obtain least-squares (adjusted) means and std. errors of these means, and performing comparisons among these means, as the LSMEANS do in SAS. I have read other messages and have looked at car, effects and Design manuals, but I am not sure if this is what I am looking for. Any clue? Felipe Dieter Menne wrote: | Felipe felipe at unileon.es writes: | | |With the SAS/STAT, I generally used the MEANS (for comparison of |arithmetic means) and the LSMEANS (for adjusted means) statements of the |GLM procedure (I think it is equivalent to lm in R). They provided a lot |of tests: LSD, Duncan, Tukey-Kramer, Bonferroni, Scheffé, SNK, etc. |However, in R I have only discovered Tukey-HSD. | | | Package multcomp with the workhorse-function simint comes close to what you | want. | | | Dieter | | __ -BEGIN PGP SIGNATURE- iEYEARECAAYFAkMpo6UACgkQWtdQtNzjBl72rwCcCuw5qxD1BWsensDI71RzhNgL MUcAnA0Iq4tfoKSr/ymIV1nEZHZijvLW =pavA -END PGP SIGNATURE- __ R-help@stat.math.ethz.ch 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] what's the best way to save global variables?
I am writing a kind of long program in R and I have some variables which I want to be globals. Where should I save them? I was thinking to create a function wich initialize all the global variables and then whenever I need them, I call this function. What if I create a file glob.R with var1-val1 var2-val2 . etc. How do I include this file in my other files/function . Is there in R some kind of include(glob.R) or something? thank you, Johan - Click here to donate to the Hurricane Katrina relief effort. [[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] Rcommander and simple chisquare
Dear John and Philippe, thanks for your replys, I finally found this menu, but I am somewhat at a loss how I should enter the observed frequencies. To take my example below, If I enter a one-column data.frame with the numbers 61 and 39, John's indicated menu is not highlighted. If I add a second column containing some factor, the menu is highlighted by I cannot select the first column. However, if I edit the data and declare the first column to be of type 'character' I can select it in the menu dialog and declare the expected frequencies, but the chisquare output doesn't make any sense. For the moment I cannot make any sense of that :-( Any help most appreciated, or a link to the tutorial/faq that explains such kind of problems. Thanks, Christian. At 11:31 -0400 15/09/05, John Fox wrote: Dear Philippe, This does a chi-square test of independence in a contingency table, not a chi-square goodness-of-fit test (which is done in the Rcmdr via Statistics - Summaries - Frequency distribution). Regards, John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Philippe Grosjean Sent: Thursday, September 15, 2005 7:32 AM To: Christian Jost Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Rcommander and simple chisquare Hello, Just look at Statistics - Contingency tables. There is an option for making the chi square test there. Best, Philippe Grosjean, ..°})) ) ) ) ) ) ( ( ( ( (Prof. Philippe Grosjean .. Christian Jost wrote: In this years biostat teaching I will include Rcommander (it indeed simplifies syntax problems that makes students frequently miss the core statistical problems). But I could not find how to make a simple chisquare comparison between observed frequencies and expected frequencies (eg in genetics where you expect phenotypic frequencies corresponding to 3:1 in standard dominant/recessif alleles). Any idea where this feature might be hidden? Or could it be added to Rcommander? Thanks, Christian. ps: in case I am not making myself clear, can Rcommander be made to perform chisq.test(c(61,39),p=c(0.75,0.25)) __ __ R-help@stat.math.ethz.ch 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] means comparison in R (post-hoc test)
Do RSiteSearch(lsmeans) and go from there. Andy From: Felipe -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Thank you, I think multcomp is very near to what I was looking for. However, I am still looking for a mean to obtain least-squares (adjusted) means and std. errors of these means, and performing comparisons among these means, as the LSMEANS do in SAS. I have read other messages and have looked at car, effects and Design manuals, but I am not sure if this is what I am looking for. Any clue? Felipe Dieter Menne wrote: | Felipe felipe at unileon.es writes: | | |With the SAS/STAT, I generally used the MEANS (for comparison of |arithmetic means) and the LSMEANS (for adjusted means) statements of the |GLM procedure (I think it is equivalent to lm in R). They provided a lot |of tests: LSD, Duncan, Tukey-Kramer, Bonferroni, Scheffé, SNK, etc. |However, in R I have only discovered Tukey-HSD. | | | Package multcomp with the workhorse-function simint comes close to what you | want. | | | Dieter | | __ -BEGIN PGP SIGNATURE- iEYEARECAAYFAkMpo6UACgkQWtdQtNzjBl72rwCcCuw5qxD1BWsensDI71RzhNgL MUcAnA0Iq4tfoKSr/ymIV1nEZHZijvLW =pavA -END PGP SIGNATURE- __ R-help@stat.math.ethz.ch 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] Rcommander and simple chisquare
Dear Christian, The Rcmdr assumes that you have a data frame with the original data, in which the variable in question is a factor. The frequency distribution is constructed for the factor. Thus, in your example, you'd have 100 observations classified on a two-category factor. What you enter directly are the hypothesized probabilities. I hope this helps, John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: Christian Jost [mailto:[EMAIL PROTECTED] Sent: Thursday, September 15, 2005 11:38 AM To: John Fox; 'Philippe Grosjean' Cc: r-help@stat.math.ethz.ch Subject: RE: [R] Rcommander and simple chisquare Dear John and Philippe, thanks for your replys, I finally found this menu, but I am somewhat at a loss how I should enter the observed frequencies. To take my example below, If I enter a one-column data.frame with the numbers 61 and 39, John's indicated menu is not highlighted. If I add a second column containing some factor, the menu is highlighted by I cannot select the first column. However, if I edit the data and declare the first column to be of type 'character' I can select it in the menu dialog and declare the expected frequencies, but the chisquare output doesn't make any sense. For the moment I cannot make any sense of that :-( Any help most appreciated, or a link to the tutorial/faq that explains such kind of problems. Thanks, Christian. At 11:31 -0400 15/09/05, John Fox wrote: Dear Philippe, This does a chi-square test of independence in a contingency table, not a chi-square goodness-of-fit test (which is done in the Rcmdr via Statistics - Summaries - Frequency distribution). Regards, John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Philippe Grosjean Sent: Thursday, September 15, 2005 7:32 AM To: Christian Jost Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Rcommander and simple chisquare Hello, Just look at Statistics - Contingency tables. There is an option for making the chi square test there. Best, Philippe Grosjean, ..°})) ) ) ) ) ) ( ( ( ( (Prof. Philippe Grosjean .. Christian Jost wrote: In this years biostat teaching I will include Rcommander (it indeed simplifies syntax problems that makes students frequently miss the core statistical problems). But I could not find how to make a simple chisquare comparison between observed frequencies and expected frequencies (eg in genetics where you expect phenotypic frequencies corresponding to 3:1 in standard dominant/recessif alleles). Any idea where this feature might be hidden? Or could it be added to Rcommander? Thanks, Christian. ps: in case I am not making myself clear, can Rcommander be made to perform chisq.test(c(61,39),p=c(0.75,0.25)) __ __ R-help@stat.math.ethz.ch 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 label a plot of a tree dendrogram with text
Dear R list, I have generated a object of class rpart in R 2.1.1 (using the wrapper function mvpart of the mvpart-package version 1.0-1). In order to label the plot of the tree dendrogram with text I used text.rpart. In this function it is possible to set the argument label to determine which values will label the nodes. In the case of my tree-object I have the following possibilities to chose the label-argument: prunedtree2$frame[1:1,] varn wt dev yval complexity ncompete nsurrogate yval2.1 yval2.2 yval2.3 yval2.4 1 EST 4258 4258 377874512 113.6452 0.23990824 1 230.56740 118.06599 78.31141 27.63598 In order to produce labels with variable n I entered: plot(prunedtree2) text.rpart(prunedtree2,label=n) Although I set label=n, it always produces labels with variable dev. I tried also to set label=yval, label=yval2.1, ... without any success, everytime it produces labels with dev. What am I doing wrong? Thanks, Pedro. __ R-help@stat.math.ethz.ch 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 vector(double, length) : vector size specified is too large....VLDs
I have what R seems to consider a very large dataset, a 12MB text file of lat,long,and height values, 130,000 rows to be exact. Here's what I get: Thomas Colson North Carolina State University Department of Forestry and Environmental Resources (919) 673 8023 [EMAIL PROTECTED] Calendar: www4.ncsu.edu/~tpcolson __ R-help@stat.math.ethz.ch 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 vector(double, length) : vector size specified is too large....VLDs
I have what R seems to consider a very large dataset, a 12MB text file of lat,long,and height values, 130,000 rows to be exact. Here's what I get: data1 - data.frame(read.table(BE3720078500WC20020828.txt,sep=,, header=T)) raw.data - as.geodata(data1) variog.1.b - variog(raw.data) variog: computing omnidirectional variogram Error in vector(double, length) : vector size specified is too large round(memory.limit()/1048576.0, 2) [1] 4000 The Vector size specified is too large seems to be a common error, but I haven't seen any workarounds posted...and the help.archive web site seems to be down. I can plot the dataset, do some elementary stats on it...no variogram though. Any ideas on how to compute variograms on datasets with 100 to 300k points? Thanks Thomas Colson North Carolina State University Department of Forestry and Environmental Resources (919) 673 8023 [EMAIL PROTECTED] Calendar: www4.ncsu.edu/~tpcolson __ R-help@stat.math.ethz.ch 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 vector(double, length) : vector size specified is too large....VLDs
At 4 GB, I'm at the 32bit windows limit. Thomas Colson North Carolina State University Department of Forestry and Environmental Resources (919) 673 8023 [EMAIL PROTECTED] Calendar: www4.ncsu.edu/~tpcolson -Original Message- From: Berton Gunter [mailto:[EMAIL PROTECTED] Sent: Thursday, September 15, 2005 2:34 PM To: 'Tom Colson' Subject: RE: [R] Error in vector(double,length) : vector size specified is too largeVLDs Any ideas on how to compute variograms on datasets with 100 to 300k points? Thanks Get more memory? ... it's cheap! :-) -- Bert Gunter Genentech __ R-help@stat.math.ethz.ch 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 vector(double, length) : vector size specified is too large....VLDs
rm(data1) variog.1.b - variog(raw.data) variog: computing omnidirectional variogram Error in vector(double, length) : vector size specified is too large Turns out I was wrong re: # of rows...it's 304,000 Same problem. Version is 2.1.1, hardware is Dual Xeon 3.6 4 GB RAM, XP Pro 64 Bit. Can reproduce the problem with 64Bit R 2.1.1 running on Fedora 4, same hardware. Thomas Colson North Carolina State University Department of Forestry and Environmental Resources (919) 673 8023 [EMAIL PROTECTED] Calendar: www4.ncsu.edu/~tpcolson -Original Message- From: Douglas Grove [mailto:[EMAIL PROTECTED] Sent: Thursday, September 15, 2005 2:23 PM To: Tom Colson Subject: Re: [R] Error in vector(double, length) : vector size specified is too largeVLDs Well you could start by removing large objects that you aren't using (e.g. 'data1') and seeing if that helps. There may be other suggestions but you haven't told us what platform you're working on, as the posting guide requests: PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Doug On Thu, 15 Sep 2005, Tom Colson wrote: I have what R seems to consider a very large dataset, a 12MB text file of lat,long,and height values, 130,000 rows to be exact. Here's what I get: data1 - data.frame(read.table(BE3720078500WC20020828.txt,sep=,, header=T)) raw.data - as.geodata(data1) variog.1.b - variog(raw.data) variog: computing omnidirectional variogram Error in vector(double, length) : vector size specified is too large round(memory.limit()/1048576.0, 2) [1] 4000 The Vector size specified is too large seems to be a common error, but I haven't seen any workarounds posted...and the help.archive web site seems to be down. I can plot the dataset, do some elementary stats on it...no variogram though. Any ideas on how to compute variograms on datasets with 100 to 300k points? Thanks Thomas Colson North Carolina State University Department of Forestry and Environmental Resources (919) 673 8023 [EMAIL PROTECTED] Calendar: www4.ncsu.edu/~tpcolson __ R-help@stat.math.ethz.ch 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] Error in vector(double, length) : vector size specified is too large....VLDs
Tom Colson [EMAIL PROTECTED] writes: rm(data1) variog.1.b - variog(raw.data) variog: computing omnidirectional variogram Error in vector(double, length) : vector size specified is too large Turns out I was wrong re: # of rows...it's 304,000 Same problem. Version is 2.1.1, hardware is Dual Xeon 3.6 4 GB RAM, XP Pro 64 Bit. Can reproduce the problem with 64Bit R 2.1.1 running on Fedora 4, same hardware. Variograms involve the differences between all pairs of points which can become a rather large number of values. 304000*303999/2 in your case, about 344GB by my reckoning. And the distances between them makes for a similar quantity. Now, some algorithms may be smarter than to keep all values in memory, but you haven't even told us where you got the variog() from. It doesn't seem to be in the standard packages, although we do have variogram() and Variogram() in spatial and nlme. -- 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] Copying from graphics window in OS X
I'm running R from an Xterm window is OSX-Tiger. Graphical windows appear as they should, but I'm having trouble copying from them--using cmd+c or the Copy option in the Edit menu won't place the graph in the clipboard (when I paste into a running OS X app, I get whatever was the last copied thing from a non-x11 window). Any ideas on how to copy from a xterm-launched graphical window? I can copy/paste into and out of the xterm command line, but I can't get anything from a graphical window. Thanks! __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Copying from graphics window in OS X
On 9/15/05 3:14 PM, Chris Wiita [EMAIL PROTECTED] wrote: I'm running R from an Xterm window is OSX-Tiger. Graphical windows appear as they should, but I'm having trouble copying from them--using cmd+c or the Copy option in the Edit menu won't place the graph in the clipboard (when I paste into a running OS X app, I get whatever was the last copied thing from a non-x11 window). Any ideas on how to copy from a xterm-launched graphical window? I can copy/paste into and out of the xterm command line, but I can't get anything from a graphical window. I don't think it is possible, but I would love to be corrected. You can simple make a png, pdf, etc. if you want to save the graphic. 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] Error in vector(double, length) : vector size specified is too large....VLDs
On 15 Sep 2005, Peter Dalgaard wrote: Tom Colson [EMAIL PROTECTED] writes: rm(data1) variog.1.b - variog(raw.data) variog: computing omnidirectional variogram Error in vector(double, length) : vector size specified is too large Turns out I was wrong re: # of rows...it's 304,000 Same problem. Version is 2.1.1, hardware is Dual Xeon 3.6 4 GB RAM, XP Pro 64 Bit. Can reproduce the problem with 64Bit R 2.1.1 running on Fedora 4, same hardware. Variograms involve the differences between all pairs of points which can become a rather large number of values. 304000*303999/2 in your case, about 344GB by my reckoning. And the distances between them makes for a similar quantity. Now, some algorithms may be smarter than to keep all values in memory, but you haven't even told us where you got the variog() from. It doesn't seem to be in the standard packages, although we do have variogram() and Variogram() in spatial and nlme. Right, this is from geoR, which uses full matrices. I think both fields and gstat can work with larger data sets. Whether model-based geostatistics is what you need for interpolating a digital elevation model is another question. -- 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] Copying from graphics window in OS X
I hope there is a way...at the moment an os X screen grab is the only way to get a quick copy. When I ran Cygwin on a PC, copying from graphic windows was as easy as ctrl+c--so it doesn't sound like an X11 limitation. I'd like to know what Cygwin was doing in the background... Sean Davis wrote: On 9/15/05 3:14 PM, Chris Wiita [EMAIL PROTECTED] wrote: I'm running R from an Xterm window is OSX-Tiger. Graphical windows appear as they should, but I'm having trouble copying from them--using cmd+c or the Copy option in the Edit menu won't place the graph in the clipboard (when I paste into a running OS X app, I get whatever was the last copied thing from a non-x11 window). Any ideas on how to copy from a xterm-launched graphical window? I can copy/paste into and out of the xterm command line, but I can't get anything from a graphical window. I don't think it is possible, but I would love to be corrected. You can simple make a png, pdf, etc. if you want to save the graphic. 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
[R] Coefficients from LM
Hi everyone, Can anyone tell me if its possibility to extract the coefficients from the lm() command? For instance, imagine that we have the following data set (the number of observations for each company is actually larger than the one showed...): Company Y X1 X2 1 y_1 x1_1x2_1 1 y_2 x1_2x2_2 1 y_3 x1_3x2_3 (...) 2 y_4 x1_4x2_4 2 y_5 x1_5x2_5 2 y_6 x1_6x2_6 (...) n y_n x1_nx2_n n y_n1x1_n1 x2_n1 n y_n2x1_n2 x2_n2 (...) I need to run a regression of Y=b0+b1*X1+b2*X2 for EACH company in the dataset and then retrieve the coefficients for each regression obtained (and t-stats and R^2) for each company and put it in another dataset/table. The procedure can be done easily done with a loop statement, but i need to retrieve each individual coefficient, t-stat, R^2, etc... I know that, using the $coefficients command will return the vector of coeffcients but I'm having trouble to assignt it to the correct row in the final dataset. Furthermore, I can't find any way of retrieving the R^2 and t-stats... Thanks for any help, Pablo. __ R-help@stat.math.ethz.ch 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] Copying from graphics window in OS X
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I'm using R for Mac OS X Aqua GUI. It allows me to copy graphs directly, simply with command-c. It's pretty nice, you can give it a try. Greetings. Felipe Chris Wiita wrote: | I'm running R from an Xterm window is OSX-Tiger. Graphical windows | appear as they should, but I'm having trouble copying from them--using | cmd+c or the Copy option in the Edit menu won't place the graph in the | clipboard (when I paste into a running OS X app, I get whatever was the | last copied thing from a non-x11 window). Any ideas on how to copy from | a xterm-launched graphical window? I can copy/paste into and out of the | xterm command line, but I can't get anything from a graphical window. | | Thanks! | | __ | R-help@stat.math.ethz.ch mailing list | https://stat.ethz.ch/mailman/listinfo/r-help | PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html | | -BEGIN PGP SIGNATURE- iEYEARECAAYFAkMp1eYACgkQWtdQtNzjBl59JgCfV7JR8kbsbYvMQG6OVt/plNTd 1S8AnRTHe6cRrwr0mxbJJGXmdkRibDvV =Pc3t -END PGP SIGNATURE- __ R-help@stat.math.ethz.ch 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] Copying from graphics window in OS X
You can try FreeSnap, a screen capture program for OS X: http://www.efritz.net/software.html Also, why are you running R from an XTerm? There is an OS X native version of R that might be better integrated with OS X for doing screen captures: http://cran.stat.ucla.edu/bin/macosx/ Jamie On 9/15/05, Chris Wiita [EMAIL PROTECTED] wrote: I'm running R from an Xterm window is OSX-Tiger. Graphical windows appear as they should, but I'm having trouble copying from them--using cmd+c or the Copy option in the Edit menu won't place the graph in the clipboard (when I paste into a running OS X app, I get whatever was the last copied thing from a non-x11 window). Any ideas on how to copy from a xterm-launched graphical window? I can copy/paste into and out of the xterm command line, but I can't get anything from a graphical window. Thanks! __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Error in vector(double, length) : vector size specified is too large....VLDs
Yes, using geoR. I can interpolate the DEM quite easily in Grass (v.surf.rst, kriging) and block kriging in ArcInfo. What we need, though, is to be able to estimate or even nail down the variogram for these data sets. Where am I going with this? I'm guessing that variables such as slope, ruggedness, etc.. are going to produce different sill, range, and nugget values, which I can then use to fine tune the interpolation process, rather than using the same spline or kriging parameters on say, a whole state boundary worth of Lidar data. And yes, I can estimate the variogram in ArcInfo (limited to 1 points) and can also import the DEM from grass into R using spgrassbut the point is to analyze the point data BEFORE I make the DEM. So I'm guessing the geoR isn't ever going to handle this size data, and I need to be using gstat? (As I write this, gstat(variogram) is plugging away for last 10 minute with no errors.) Thanks for quick replies Thomas Colson North Carolina State University Department of Forestry and Environmental Resources (919) 673 8023 [EMAIL PROTECTED] Calendar: www4.ncsu.edu/~tpcolson -Original Message- From: Roger Bivand [mailto:[EMAIL PROTECTED] Sent: Thursday, September 15, 2005 3:28 PM To: Peter Dalgaard Cc: Tom Colson; r-help@stat.math.ethz.ch Subject: Re: [R] Error in vector(double, length) : vector size specified is too largeVLDs On 15 Sep 2005, Peter Dalgaard wrote: Tom Colson [EMAIL PROTECTED] writes: rm(data1) variog.1.b - variog(raw.data) variog: computing omnidirectional variogram Error in vector(double, length) : vector size specified is too large Turns out I was wrong re: # of rows...it's 304,000 Same problem. Version is 2.1.1, hardware is Dual Xeon 3.6 4 GB RAM, XP Pro 64 Bit. Can reproduce the problem with 64Bit R 2.1.1 running on Fedora 4, same hardware. Variograms involve the differences between all pairs of points which can become a rather large number of values. 304000*303999/2 in your case, about 344GB by my reckoning. And the distances between them makes for a similar quantity. Now, some algorithms may be smarter than to keep all values in memory, but you haven't even told us where you got the variog() from. It doesn't seem to be in the standard packages, although we do have variogram() and Variogram() in spatial and nlme. Right, this is from geoR, which uses full matrices. I think both fields and gstat can work with larger data sets. Whether model-based geostatistics is what you need for interpolating a digital elevation model is another question. -- 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] Copying from graphics window in OS X
I just tried using Ctrl-C to do a copy a plot from a graphic window using the Cocoa version of OS X that you can download from the link below and I was able to paste the plot into a document. Jamie On 9/15/05, Jamieson Cobleigh [EMAIL PROTECTED] wrote: Also, why are you running R from an XTerm? There is an OS X native version of R that might be better integrated with OS X for doing screen captures: http://cran.stat.ucla.edu/bin/macosx/ __ R-help@stat.math.ethz.ch 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] Copying from graphics window in OS X
I'm connecting over SSH to a linux server to do calculations, so I need to be able to run the remote linux console on the Mac. Jamieson Cobleigh wrote: You can try FreeSnap, a screen capture program for OS X: http://www.efritz.net/software.html Also, why are you running R from an XTerm? There is an OS X native version of R that might be better integrated with OS X for doing screen captures: http://cran.stat.ucla.edu/bin/macosx/ Jamie On 9/15/05, Chris Wiita [EMAIL PROTECTED] wrote: I'm running R from an Xterm window is OSX-Tiger. Graphical windows appear as they should, but I'm having trouble copying from them--using cmd+c or the Copy option in the Edit menu won't place the graph in the clipboard (when I paste into a running OS X app, I get whatever was the last copied thing from a non-x11 window). Any ideas on how to copy from a xterm-launched graphical window? I can copy/paste into and out of the xterm command line, but I can't get anything from a graphical window. Thanks! __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Converting coordinates to actual distances
Paul Brewin wrote: Hello, I've been searching for a method of converting Lat/Lon decimal coordinates into actual distances between points, and taking into account the curvature of the earth. Is there such a package in R? I've looked at the GeoR package, but this does not seem to contain what I am looking for. Ideally the output would be a triangular matrix of distances. Hi Paul, Below is an implementation of the Haversine formula that I once had to use for calculating distances between telephone exchanges. It's in C, but could be translated to R fairly easily. I've included a degrees to radians conversion as well. Jim #define APOSTROPHE 39 #define EARTH_RADIUS 6367 typedef struct { char *basename; char *number[2]; double lat[2]; double lon[2]; char delimiter[4]; char *data02; char *data03; char *data07; char *data08; FILE *input; FILE *output; } PSTS_INFO; /* DegreesToRadians Converts a string of the form nnndnn'nn[NSEW] to a value in radians. East (E) and South (S) are arbitrarily negative. Note that this routine will generate garbage if not passed a string of the correct form, although it is unlikely to do too much damage. Return valuevalue of the string in radians -10 input not correct format */ double DegreesToRadians(char *dms) { int index = 0; double degrees; double minutes = 0; double seconds = 0; int mult; degrees = atof(dms); while(isdigit(dms[index])) index++; if(dms[index++] == 'd') { minutes = atof(dms[index]); minutes /= 60; while(isdigit(dms[index])) index++; if(dms[index++] == APOSTROPHE) { seconds = atof(dms[index]); seconds /= 360; while(isdigit(dms[index]) dms[index]) index++; if(dms[index++] == QUOTES) { mult = (dms[index] == 'S' || dms[index] == 'E') ? -1 : 1; degrees += minutes + seconds; return(mult * degrees * M_PI/180); } } } return(-10); } /* GreatCircleDistance (Formula and recommendations for calculation taken from: http://www.census.gov/cgi-bin/geo/gisfaq?Q5.1 by Bob Chamberlain - [EMAIL PROTECTED]) Calculates 'great circle' distances using the Haversine formula, based upon a spherical earth of EARTH_RADIUS radius. This version extracts the two latitude/longitude pairs (as radians) from the PSTS_INFO structure. */ double GreatCircleDist(PSTS_INFO *psts_info) { double a,d,dlon,dlat; dlon = psts_info-lon[1] - psts_info-lon[0]; dlat = psts_info-lat[1] - psts_info-lat[0]; a = pow(sin(dlat / 2),2) + cos(psts_info-lat[0]) * cos(psts_info-lat[1]) * pow(sin(dlon / 2),2); /* The penultimate result may be calculated either way - the first is bulletproof, the second is a bit faster */ d = 2 * atan2(sqrt(a),sqrt(1 - a)); /* d = 2 * asin(sqrt(a));*/ return(d * EARTH_RADIUS); } __ R-help@stat.math.ethz.ch 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] Coefficients from LM
Tsk, tsk. You don't seem to be looking very hard. Here's an example with a glm; lm() works the same way but has fewer internal objects. mod3 - glm(tree ~ altitude, family = binomial) You can use names() to find out what's inside: names(mod3) [1] coefficients residuals fitted.values effects R [6] rank qrfamily linear.predictors deviance [11] aic null.deviance iter weights prior.weights [16] df.residual df.null y converged boundary [21] model call formula terms data [26] offsetcontrol methodcontrasts xlevels There are lots of ways to retrieve the parameter estimates: coefficients(mod3) (Intercept)altitude 13.43360163 -0.01220884 mod3$coeff (Intercept)altitude 13.43360163 -0.01220884 mod3$coef[2] altitude -0.01220884 as.numeric(mod3$coef[2]) [1] -0.01220884 Charles Annis, P.E. [EMAIL PROTECTED] phone: 561-352-9699 eFax: 614-455-3265 http://www.StatisticalEngineering.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Pablo Gonzalez Sent: Thursday, September 15, 2005 4:09 PM To: r-help@stat.math.ethz.ch Subject: [R] Coefficients from LM Hi everyone, Can anyone tell me if its possibility to extract the coefficients from the lm() command? For instance, imagine that we have the following data set (the number of observations for each company is actually larger than the one showed...): Company Y X1 X2 1 y_1 x1_1x2_1 1 y_2 x1_2x2_2 1 y_3 x1_3x2_3 (...) 2 y_4 x1_4x2_4 2 y_5 x1_5x2_5 2 y_6 x1_6x2_6 (...) n y_n x1_nx2_n n y_n1x1_n1 x2_n1 n y_n2x1_n2 x2_n2 (...) I need to run a regression of Y=b0+b1*X1+b2*X2 for EACH company in the dataset and then retrieve the coefficients for each regression obtained (and t-stats and R^2) for each company and put it in another dataset/table. The procedure can be done easily done with a loop statement, but i need to retrieve each individual coefficient, t-stat, R^2, etc... I know that, using the $coefficients command will return the vector of coeffcients but I'm having trouble to assignt it to the correct row in the final dataset. Furthermore, I can't find any way of retrieving the R^2 and t-stats... Thanks for any help, Pablo. __ R-help@stat.math.ethz.ch 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] Nolinear mixed-effects models (nlme)
Do you send information about lactation curve analyse with no linear mixed model, with fixed effects (herd, year season, parity) and random effects (cow)?. Than you very much Mario Fernando [[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] about cutree
On Sep 15, 2005, at 10:26 AM, Baoqiang Cao wrote: Hi Everyone, I'm trying to use cutree to get the clusters after hclust. What I used is: mycluster-cutree(cnclust,h=0.5) Now, my problem is, how can I get the actual clusters? Thanks! Best, Baoqiang Cao Doesn't print(mycluster) give you the clusters? You could use something like: R split(names(mycluster), mycluster) You may need to better define get the actual clusters to get a better answer. Stephen __ R-help@stat.math.ethz.ch 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] what's the best way to save global variables?
Johan, On Sep 15, 2005, at 11:39 AM, johan Faux wrote: I am writing a kind of long program in R and I have some variables which I want to be globals. Where should I save them? I was thinking to create a function wich initialize all the global variables and then whenever I need them, I call this function. In most cases, you would write a function that would return an object of class list, the components of which would be the values you want to use later. For example myfun - function([stuff]){ [stuff] return(list = (var1 = val1, var2 = val2) } and your call would be glob - myfun([stuff]) and you would access val1 with glob$var1 What if I create a file glob.R with var1-val1 var2-val2 . etc. How do I include this file in my other files/function . Is there in R some kind of include(glob.R) or something? thank you, Johan Use source(glob.R) Check out An introduction to R (http://cran.r-project.org/doc/manuals/R-intro.html). Stephen __ R-help@stat.math.ethz.ch 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] Searchable archives
I cannot access the searchable archives at www.tolstoy.newcastle.au/~rking/R. Does anyone else have this problem? -- Fiona H. Evans http://www.cmis.csiro.au/Fiona.Evans [[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] Searchable archives
Yes, I've had the same trouble. Robert may be able to sort this out. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, 16 September 2005 3:34 PM To: r-help@stat.math.ethz.ch Subject: [R] Searchable archives I cannot access the searchable archives at www.tolstoy.newcastle.au/~rking/R. Does anyone else have this problem? -- Fiona H. Evans http://www.cmis.csiro.au/Fiona.Evans [[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