Re: [R] How to preserve date format while aggregating

This is completely wrong: min _is_ defined for date-times: min(.leap.seconds) [1] "1972-07-01 01:00:00 BST" Please do study the posting guide and do your homework before posting: you seem unaware of what the POSIXct class is, so ?DateTimeClasses is one place you need to start. And method

[R] Memory allocation problem (during kmeans)

Dear all, I am trying to apply kmeans clusterring on a data file (size is about 300 Mb) I read this file using x=read.table('file path' , sep=" ") then i do kmeans(x,25) but the process stops after two minutes with an error : Error: cannot allocate vector of size 907.3 Mb when i read the arc

[R] Compiling date

Hi, I have following kind of dataset (all are dates) in my Excel sheet. 09/08/08 09/05/08 09/04/08 09/02/08 09/01/08 29/08/2008 28/08/2008 27/08/2008 26/08/2008 25/08/2008 22/08/2008 21/08/2008 20/08/2008 18/08/2008 14/08/2008 13/08/2008 08/12/08 08/11/08 08/08/08 08/07/08 However I want to use

[R] match problem by rownames

Hi all, While dat['a1',] and dat['a10',] produce the same results in the following example, I'd like dat['a1',] to return NAs. dat <- data.frame(x1 = paste(letters[1:5],10, sep=''), x2=rnorm(5)) rownames(dat) <- dat$x1 dat['a1',] dat['a10',] > sessionInfo() R version 2.7.2 (2008-08-25) i386-pc-m

Re: [R] yahoo finance into R

thomastos wrote: Hi R, I am familiar with the basics of R. To learn more I would like how to get data from Yahoo!finance directly into R. So basically I want a data frame or matrix to do some data analysis. How do I do this? RSiteSearch("yahoo") get.hist.quote() from tseries yahooSeries()

Re: [R] cluster/snow question

Hi Tolga, in SNOW you have to start a cluster with the command > library(snow) > cluster <- makeCluster(#nodes) The object "cluster" is a list with an object for each node and each object again is a list with all informations (rank, comm, tags) The size of the cluster is the length of the lis

Re: [R] Splitting Data Frame into Two Based on Source Array

data_main[ match(src,data_main$V1), ] and the compliment of src (call it srcc) data_main[ match(srcc,data_main$V1), ] ...this only works so long as there is only one occurrance of each item in V1 in V1. --Adam On Tue, 9 Sep 2008, Gundala Viswanath wrote: Dear all, Suppose I have this data

[R] Overlapping Area

Hi, I'm calculating the KL Divergence measure between two distributions and like to know the KLD measure corresponding to an overlap of certain value(example 0.35). Is there any package in R which calculates the overlapping area between two densities/curves. Thanks -- View this message in cont

[R] yahoo finance into R

Hi R, I am familiar with the basics of R. To learn more I would like how to get data from Yahoo!finance directly into R. So basically I want a data frame or matrix to do some data analysis. How do I do this? Thank you very much. Thomas -- View this message in context: http://www.nabble.com/

Re: [R] Splitting Data Frame into Two Based on Source Array

Try this which gives a list of two data frames: split(data_main, data_main$V1 %in% c("foo", "bar")) On Mon, Sep 8, 2008 at 11:17 PM, Gundala Viswanath <[EMAIL PROTECTED]> wrote: > Dear all, > > Suppose I have this data frame: > > >> data_main > V1 V2 > foo13.1 > bar 12.0 > qux 10.4 >

Re: [R] Splitting Data Frame into Two Based on Source Array

Dear Gundala, Try this: # Data set data_main=read.table(textConnection("V1 V2 foo13.1 bar 12.0 qux 10.4 cho 20.33 pox 8.21"),header=TRUE) attach(data_main) closeAllConnections() # Criteria src=c('bar','pox') # data_child1 data_main[V1%in%src,] # data_child2_complement data_main[!V1

[R] Splitting Data Frame into Two Based on Source Array

Dear all, Suppose I have this data frame: > data_main V1 V2 foo13.1 bar 12.0 qux 10.4 cho 20.33 pox 8.21 And I want to split the data into two parts first part are the one contain in the source array: > src [1] "bar" "pox" and the other one the complement. In the end we hope t

Re: [R] densities with overlapping area of 0.35

Just a correction: if we take X+2a then everything is OK (the curves intersect at a), so a = 0.9345893 is correct but one must take X ~ N(0,1) and Y ~N(2*a,1). --- On Tue, 9/9/08, Moshe Olshansky <[EMAIL PROTECTED]> wrote: > From: Moshe Olshansky <[EMAIL PROTECTED]> > Subject: Re: [R] densities

Re: [R] densities with overlapping area of 0.35

Let X be normally distributed with mean 0 and let f be it's density. Now the density of X+a will be f shifted right by a. Since the density is symmetric around mean it follows that the area of overlap of the two densities is exactly P(X>a) + P(X<-a). So if X~N(0,1), we want P(X>a) + P(X<-a) = 2P

Re: [R] isolate elements in vector that match one of many possible values

On 9/09/2008, at 1:55 PM, Andrew Barr wrote: Hi all, I want to get the index numbers of all elements of a vector which match any of a long series of possible values. Say x <- c(1,2,3,4) and I want to know which values are equal to 1, 2 or 4. I could do which(x == 1 | x==2 | x==4) [1] 1

[R] densities with overlapping area of 0.35

Hi, I like to generate two normal densities such that the overlapping area between them is 0.35. Is there any code/package available in R to do that?? Regards, Lavan -- View this message in context: http://www.nabble.com/densities-with-overlapping-area-of-0.35-tp19384741p19384741.html Sent fr

Re: [R] make methods work in lapply - remove lapply's environment

On 08/09/2008 9:37 PM, Tim Hesterberg wrote: I've defined my own version of summary.default, that gives a better summary for highly skewed vectors. If I call summary(x) the method is used. If I call summary(data.frame(x)) the method is not used. I've traced this to lapply; this uses the ne

[R] isolate elements in vector that match one of many possible values

Hi all, I want to get the index numbers of all elements of a vector which match any of a long series of possible values. Say x <- c(1,2,3,4) and I want to know which values are equal to 1, 2 or 4. I could do which(x == 1 | x==2 | x==4) [1] 1 2 4 This gets really ugly though, when the list of v

[R] make methods work in lapply - remove lapply's environment

I've defined my own version of summary.default, that gives a better summary for highly skewed vectors. If I call summary(x) the method is used. If I call summary(data.frame(x)) the method is not used. I've traced this to lapply; this uses the new method: lapply(list(x), function(x) summary

[R] naive variance in GEE

Hi, The standard error from logistic regression is slightly different from the naive SE from GEE under independence working correlation structure. Shouldn't they be identical? Anyone has insight about this? Thanks, Qiong a<-rbinom(1000,1) b<-rbinom(1000,2,0.1) c<-rbinom(1000,10,0.5) summary(

Re: [R] R_USER - in which file should I include it?

Try adding this at the end of your etc/Rprofile.site file. That file should already be there so you don't have to create it, just edit it. cat("Hello from Rprofile.site\n") setwd("C:/Users/eduardo/Documents") You may need to edit it as Administrator. You should see the Hello message in which c

Re: [R] R_USER - in which file should I include it?

Hello I am not sure whether R starts from the same dir. For instance: a) if I double-click on R-2.7.2 icon and then issue the command getwd(), the result is: > getwd() [1] "C:/Users/eduardo/Documents" b) If R starts from within Tinn-R, the result is: getwd() [1] "C:/Program Files/R/R-2/bin"

Re: [R] R_USER - in which file should I include it?

Could you explain more clearly what you mean by "the same"? Do you mean that each time you click on R 2.7.2 icon on your desktop that running this from the R console: getwd() is the same directory on each startup? Isn't that already the case? I don't think you need to set any environment variabl

Re: [R] R_USER - in which file should I include it?

I don't know anything about vista, but the libraries for windows XP are installed in the same place for each version there is a parallel installation. This is by design. sorry I can't be of more help On Mon, Sep 8, 2008 at 7:10 PM, Eduardo M. A. M.Mendes <[EMAIL PROTECTED]> wrote: > Hello > > I a

[R] R_USER - in which file should I include it?

Hello I am a newbie. I had my R upgraded from 2.7.1 to 2.7.2 and in doing so I decided to install all 2.7 versions under c:\program files\R\2.7 from now on (2.7.1 is located under .\2.7.1) Although I don't like the idea (I am running Vista), I have edited etc\Renviron.site to contain: R_USER=

[R] correct lme syntax for this problem?

Hello all, I am about to send off a manuscript and, although I am fairly confident I have used the lme function correctly, I want to be 100% sure. Could some kind soul out there put my mind at ease? I am simply interested in whether a predictor (SPI) is related to height. However, there are five

Re: [R] Question about multiple regression

Hi Dimitri, On Mon, 8 Sep 2008, Dimitri Liakhovitski wrote: Dear R-list, maybe some of you could point me in the right direction: Are you aware of any FREE Fortran or Java libraries/actual pieces of code that are VERY efficient (time-wise) in running the regular linear least-squares multiple r

Re: [R] How to preserve date format while aggregating

min does work for POSIXct and Date too: > ct <- ISOdatetime(2008, 1, 1:10, 0, 0, 0) > min(ct) [1] "2008-01-01 EST" > min(as.Date(ct)) [1] "2008-01-01" On Mon, Sep 8, 2008 at 6:57 PM, Adam D. I. Kramer <[EMAIL PROTECTED]> wrote: > Hi Erich, > > Since min() is defined for numbers and not dates,

Re: [R] Saving functions

Hi Robin, source("filename.R") will open filename.R from the working directory and behave as if you had typed in its contents, line by line. You can include a full path if you like, or use file.choose() for a file not in your working directory. --Adam On Mon, 8 Sep 2008, Williams, Robin wrote

Re: [R] How to preserve date format while aggregating

Hi Erich, Since min() is defined for numbers and not dates, the problem is in the min() function. min() is converting from date format to number format. Your best bet is to make this conversion explicit...such that it is reversable. So, convert the date into UTC, then UTC to seconds since epoch,

Re: [R] Request for advice on character set conversions (those damn Excel files, again ...)

Emmanuel Charpentier wrote: On Mon, 08 Sep 2008 01:45:51 +0200, Peter Dalgaard wrote : Emmanuel Charpentier wrote: Dear list, [ Snip ... ] This looks reasonably sane, I think. The last loop could be d[] <- lapply(d, conv1, from, to), but I think that is cosmetic.

Re: [R] extracting max row from data matrix

Hi Srini, This may be as simple as tapply(weight,fruit,max) or t(that) if you want it as you specified. --Adam On Sun, 7 Sep 2008, Srinivas Iyyer wrote: dear group, i have a data matrix with some replicate items with different values. I want to extract the row with max value. for example:

Re: [R] ON MAC, how to copy a plot on to Word document?

word 2004 post script (plot created with postscript()) does not work but pdf graphics work on the mac. I suggest not copying and pasting because, in my expreience, the output is not as good as pdf et al. On Mon, Sep 8, 2008 at 6:19 PM, Frank E Harrell Jr <[EMAIL PROTECTED]> wrote: > Word in Offi

Re: [R] nested ANOVA with fixed and random variables missing data

On Mon, 8 Sep 2008, [EMAIL PROTECTED] wrote: I have a nested ANOVA, with a fixed factor "tmt" nested within "site" (random). There are missing values in the data set. aeucs, tmt and site have been defined as objects I have tried: model1=lme(aeucs~tmt,random=~1|tmt/site) I think you want lm

Re: [R] how to draw a vertical line from points to x-axis

I think you want the ?lines function. To connect a point (x,y) to the x-axis, lines(x=c(x,x),y=c(y,0)) ...draws a line from that point to the x-axis. You may also want to specify pch=c(?,""),type="b" where ? is the original point type (which you don't want to "run over") and "" is the pch for t

Re: [R] ON MAC, how to copy a plot on to Word document?

Word in Office 2003 (and I assume Office 2007) can insert encapsulated postscript with no problems on Windows, and display the graphic fine. Does this not work with Mac? Doesn't dragging a file into Word lose the full resolution of postscript? Frank Adam D. I. Kramer wrote: For what it's w

Re: [R] Averaging 'blocks' of data

Hi Steve, You probably want to check out ?by or ?aggregate, maybe using (rownames(df) %/% 60) : (colnames(df) %/% 60) as your index variable. --Adam On Sun, 7 Sep 2008, Steve Murray wrote: Dear all, I have a large dataset which I hope to reduce in size, to make it more useable. I h

Re: [R] ON MAC, how to copy a plot on to Word document?

For what it's worth, copy/paste between R and Word 2008 works perfectly. It doesn't work at all for Word 2004...so that's at least ONE improvement (the only one I've seen) in versions of Word. --Adam On Sun, 7 Sep 2008, Stefan Evert wrote: On 7 Sep 2008, at 16:57, John Kane wrote: I think yo

Re: [R] Test for equality of complicatedly related average correlations

Hi Ralph, My approach provides the same answer by asking a different question. Effectively, my approach tests whether the difference between timepoints is larger for X than for Y (and also gives you the main effects of whether X has a higher mean than Y and whether scores increase or decrease ove

Re: [R] Time Series x-axis labeling

Try this: plot(tsData, ylab="Values", xlab="Zeit", xaxt = "n") axis(1, time(tsData), rep(substr(month.abb, 1, 1), length = length(tsData)), cex.axis = .3, tcl = -.5) jan <- time(tsData)[cycle(tsData) == 1] # january axis(1, jan, FALSE, tcl = -1) abline(v = jan, lty = 2) and also look at

[R] Gumbell distribution - minimum case

Hello, I would like to sample from a Gumbell (minimum) distribution. I have installed package {evd} but the Gumbell functions there appear to refer to the maximum case. Unfortunately, setting the scale parameter negative does not appear to work. Is there a separate package for the Gumbell

Re: [R] Request for advice on character set conversions (those damn Excel files, again ...)

On Mon, 08 Sep 2008 01:45:51 +0200, Peter Dalgaard wrote : > Emmanuel Charpentier wrote: >> Dear list, >> [ Snip ... ] > This looks reasonably sane, I think. The last loop could be d[] <- > lapply(d, conv1, from, to), but I think that is cosmetic. You can't > really do much better becau

Re: [R] Question about multiple regression

Thanks a lot, everybody! On Mon, Sep 8, 2008 at 3:11 PM, Lucke, Joseph F <[EMAIL PROTECTED]> wrote: > Although I along with the other believe there probably is an efficient R > solution, the answer to your direct question can perhaps be found at > http://www.fortran.com/. The free GNU G95 fortra

[R] Time Series x-axis labeling

Hello, how is it possible to plot a time series of monthly data over several years such that the x-axis shows the first letter of the month and displays a grid line at every year? I am new to R and had no real success until now. I have: > library(utils) > oneMonth = 1.0 / 12.0 > tsData = ts(rnorm

Re: [R] Read from url requiring authentication?

Damien schrieb: I'm looking into opening an url on a server which requires authentication. Under a Windows Operating System you could try to start R with the --internet2 option. This worked in my case. Best regards Rene __ R-help@r-project.org ma

[R] Forecast for stl?

Once the series has been decomposed into the seasonal, trend, and remainder components with stl is there a 'predict' like function that I can use this data to forecast the next 'n' values? I didn't see one in the stl documentation. So if such a function does not exist I was wondering if I could

[R] cluster/snow question

Dear R Users, I am attempting to use the snow package for clustering. Is there a way to identfy, in the environment of each node, a rank for that node and also, the total size of the cluster ? By way of analogy, I am looking for the functions in snow equivalent to mpi.comm.rank() and mpi.comm

Re: [R] Pie chart and labels

On Monday, 08 September 2008, 15:24 (UTC-0400), polishookm wrote: > With a pie chart > > pie(c(11, 15, 16, 29, 31)) > > how can I generate labels for the chart, such as > > orange: 11 > green: 15 > blue: 16 > red: 29 > purple: 31 > pie(c(11, 15, 16, 29, 31),labels=c("what","ever","you","want","d

Re: [R] Pie chart and labels

Thanks Jorge, that's perfect. Jorge Ivan Velez wrote: Dear Mark, See argument "labels" in ?pie. x=c(11, 15, 16, 29, 31) pie(x,labels=x,col=c('orange','green','blue','red','purple')) HTH, Jorge On Mon, Sep 8, 2008 at 3:24 PM, polishookm <[EMAIL PROTECTED]

Re: [R] Modulous operator?

Thank you. I definitely did not want Mod. Kevin Peter Dalgaard <[EMAIL PROTECTED]> wrote: > Csardi Gabor wrote: > > ?"%%" > > > Yup. Notice, by the way, that "modulus" [sic] is ambiguous: > > > Mod(1+1i) > [1] 1.414214 > > > 46 %% 7 > [1] 4 > > > > Gabor > > > > On Mon, Sep 08, 20

Re: [R] Pie chart and labels

Dear Mark, See argument "labels" in ?pie. x=c(11, 15, 16, 29, 31) pie(x,labels=x,col=c('orange','green','blue','red','purple')) HTH, Jorge On Mon, Sep 8, 2008 at 3:24 PM, polishookm <[EMAIL PROTECTED]>wrote: > With a pie chart > > pie(c(11, 15, 16, 29, 31)) > > how can I generate labels

[R] Pie chart and labels

With a pie chart pie(c(11, 15, 16, 29, 31)) how can I generate labels for the chart, such as orange: 11 green: 15 blue: 16 red: 29 purple: 31 rather than the default labels that are drawn automatically 1 2 3 4 5 -- Mark Polishook, D.M.A. Technology Coordinator Morehead Hall, Rm 132 Center fo

Re: [R] Modulous operator?

Csardi Gabor wrote: ?"%%" Yup. Notice, by the way, that "modulus" [sic] is ambiguous: > Mod(1+1i) [1] 1.414214 > 46 %% 7 [1] 4 Gabor On Mon, Sep 08, 2008 at 11:58:45AM -0700, [EMAIL PROTECTED] wrote: Is there a mod (like C '%' operator) operator in 'R'? I tried to help.search("mod

Re: [R] Read from url requiring authentication?

On Mon, 8 Sep 2008, Damien wrote: Hi all, I'm looking into opening an url on a server which requires authentication. After failing to find some kind of connection structure to fill in I turned to explicitly stating the credentials in the url itself (e.g. "http://username:[EMAIL PROTECTED]"). S

Re: [R] Question about multiple regression

Although I along with the other believe there probably is an efficient R solution, the answer to your direct question can perhaps be found at http://www.fortran.com/. The free GNU G95 fortran compiler is at http://www.g95.org/ Joe -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL

Re: [R] Modulous operator?

?"%%" Gabor On Mon, Sep 08, 2008 at 11:58:45AM -0700, [EMAIL PROTECTED] wrote: > Is there a mod (like C '%' operator) operator in 'R'? I tried to > help.search("mod") and there were too many hits for the query to be useful. > > Kevin > > __ > R-help@

[R] Modulous operator?

Is there a mod (like C '%' operator) operator in 'R'? I tried to help.search("mod") and there were too many hits for the query to be useful. Kevin __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the post

Re: [R] Question about multiple regression

R squared is: 1 - sum(residuals^2)/crossprod(y - mean(y)) On Mon, Sep 8, 2008 at 2:27 PM, Dimitri Liakhovitski <[EMAIL PROTECTED]> wrote: > I could get an r squared from lm.fit by correlating fitted.values and > my response variable. > But could I do it somehow using Sums of Squares? I am clear

Re: [R] RMPI Question

Dear Martin, Many thanks. This is very useful, much appreciated. Regards, Tolga Martin Morgan <[EMAIL PROTECTED]> 08/09/2008 19:08 To [EMAIL PROTECTED] cc r-help@r-project.org Subject Re: [R] RMPI Question Hi Tolga -- [EMAIL PROTECTED] writes: > Dear R Users, > > I have a relatively

Re: [R] Question about multiple regression

I could get an r squared from lm.fit by correlating fitted.values and my response variable. But could I do it somehow using Sums of Squares? I am clear on SS for residuals. But where is SS for the model or the total SS in lm.fit output? Thank you! Dimitri On Mon, Sep 8, 2008 at 1:57 PM, Gabor Grot

Re: [R] Beginner graphics device questions.

> On Monday, 08 September 2008, 10:59 (UTC-0700), > [EMAIL PROTECTED] wrote: > If I issue another 'plot' command will it also be redirected > to the 'png' device? As long as that png device is the currently active device, yes. > Does dev.off essentially reset the graphics > device to the de

[R] mixed model MANCOVA

Hello, I need to perform a mixed-model (with nesting) MANCOVA, using Type III sums of squares. I know how to perform each of these types of tests individually, but I am not sure if performing a mixed-model MANCOVA is possible. Please let me know. Erika <>< <>< <>< <><

[R] question on extracting binary data from pdf image

Dear List, If I use the following code to generate a pdf, pdf(filename) image(x, y, !is.na(z), col=c("green","black")) dev.off() Can someone extract the 0-1 data (that is, is.na(z)) from the Adobe source? Thank you. Xiaochun [[alternative HTML version deleted]] _

Re: [R] RMPI Question

Hi Tolga -- [EMAIL PROTECTED] writes: > Dear R Users, > > I have a relatively simple rmpi question. > > My configation is: > - R version 2.7.2 > - rmpi 0.5-6 > - Deino MPI 1.1.0 > - Windows XP SP2 > > After succesfully spawning slaves, I am trying to assign values to > variables in the environm

Re: [R] Articles about comparision between R and others softwares

Hi, There are more articles, papers, contents about statistical tools ??? (Perfomance, packages, Usability, difficulty in handling the software. something) Thank You Ricardo -- View this message in context: http://www.nabble.com/Articles-about-comparision-between-R-and-others-softwares-tp

[R] mixed model MANCOVA

Hello, I need to perform a mixed-model (with nesting) MANCOVA, using Type III sums of squares. I know how to perform each of these types of tests individually, but I am not sure if performing a mixed-model MANCOVA is possible. Please let me know. Erika <>< <>< <>< <>< <><

[R] Read from url requiring authentication?

Hi all, I'm looking into opening an url on a server which requires authentication. After failing to find some kind of connection structure to fill in I turned to explicitly stating the credentials in the url itself (e.g. "http://username:[EMAIL PROTECTED]"). Sadly this didn't do the trick either

Re: [R] Sort (indices only)

?order This will give you the indices that you can use. On Mon, Sep 8, 2008 at 1:38 PM, <[EMAIL PROTECTED]> wrote: > I am using the function 'spectrum'. It returns two arrays that are > interesting to me. One would be the wieght or density of a given frequency > with the irequency given in ano

[R] Beginner graphics device questions.

Please forgive me if this has been asked before but I could not readily find an answer. First, from the example I was able to determine that graphics device commands such as plot can be redirected using 'png' and 'bmp' type commands. I tried this and it works as I understand it. The question is

Re: [R] Question about multiple regression

On Mon, Sep 8, 2008 at 1:47 PM, Dimitri Liakhovitski <[EMAIL PROTECTED]> wrote: > Thank you everyone for your responses. I'll answer several questions. > > 1. > Disclaimer: I have **NO IDEA** of the details of what you want > to do or why >> -- but I am willing to bet that there are better ways of

Re: [R] Question about multiple regression

Thank you everyone for your responses. I'll answer several questions. 1. > Disclaimer: I have **NO IDEA** of the details of what you want to do or why > -- but I am willing to bet that there are better ways of doing it than 1.8 > mm multiple refressions that take 270 secs each!! (which I find di

Re: [R] (no subject)

Dear Jason, If I understood, you are looking for a way of represent a response surface. I know that there are other very interesting (and most indicated) solutions, but give a look at the akima package Best wishes, miltinho astronauta brazil === x<-runif(100) y<-rnorm(100) binvar<-c(rep(1,70)

[R] Sort (indices only)

I am using the function 'spectrum'. It returns two arrays that are interesting to me. One would be the wieght or density of a given frequency with the irequency given in another array. I would like to take the top 'n' weights which would be the top 'n' frequencies contributing to the signal. One

Re: [R] Question about multiple regression

Try: sum(lm.fit(x, y)$residuals^2) On Mon, Sep 8, 2008 at 12:52 PM, Dimitri Liakhovitski <[EMAIL PROTECTED]> wrote: > Thank you for reminding me, Gabor. I forgot to mention: So far, I have > run one test set of regressions using lm. It took R 270 sec. I need to > run 1,800,000 of those, which wou

Re: [R] how to draw a vertical line from points to x-axis

Anny, You can also do the following plot(0:10, 0:10, pch=16, type="h") Cheers../Murli -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Peter Alspach Sent: Sunday, September 07, 2008 5:10 PM To: Anny Huang; r-help@r-project.org Subject: Re: [R] how to dra

Re: [R] Plot of large dataset

I'd start with scatterplots of the two subsets (pass vs fail), but with 280k points, those are likely to be fairly uninformative masses of black ink). However, there might be enough separation between them that you don't need anything else. If not, then a pair of hexbin plots (from the Biocon

Re: [R] Question about multiple regression

Disclaimer: I have **NO IDEA** of the details of what you want to do or why -- but I am willing to bet that there are better ways of doing it than 1.8 mm multiple refressions that take 270 secs each!! (which I find difficult to believe in itself -- are you sure you are doing things right? Somethi

[R] fft: characteristic function to distribution

Hello members! This question had been posted by Thomas Steiner in May, but I couldn't locate any followups thereafter, see: https://stat.ethz.ch/pipermail/r-help/2008-May/161483.html I want to raise up this question in a (hopefully) even simpler way: Given a random variable X, it's characteristi

[R] RMPI Question

Dear R Users, I have a relatively simple rmpi question. My configation is: - R version 2.7.2 - rmpi 0.5-6 - Deino MPI 1.1.0 - Windows XP SP2 After succesfully spawning slaves, I am trying to assign values to variables in the environment of each slave and run some simple calcs. This appears to

Re: [R] Question about multiple regression

Yes, see my previous e-mail on how long R takes (270 seconds for one of the 1,800,000 sets I need) - using system.time. Not sure how to test the same for Fortran... On Mon, Sep 8, 2008 at 12:51 PM, Prof Brian Ripley <[EMAIL PROTECTED]> wrote: > Are you sure R's ways are not fast enough (there are

Re: [R] Question about multiple regression

Thank you for reminding me, Gabor. I forgot to mention: So far, I have run one test set of regressions using lm. It took R 270 sec. I need to run 1,800,000 of those, which would imply 15.4 years of computing time :) I have not done the same for lm.fit because I am not sure how to get model R squar

Re: [R] Question about multiple regression

Are you sure R's ways are not fast enough (there are many layers underneath lm)? For an example of how you might do this at C/Fortran level, see the function lqs() in MASS. On Mon, 8 Sep 2008, Dimitri Liakhovitski wrote: Dear R-list, maybe some of you could point me in the right direction:

Re: [R] question on extracting binary data from pdf image

On Mon, 8 Sep 2008, Li, Xiaochun wrote: Dear List, If I use the following code to generate a pdf, pdf(filename) image(x, y, !is.na(z), col=c("green","black")) dev.off() Can someone extract the 0-1 data (that is, is.na(z)) from the Adobe source? Yes (from the PDF file, which is not due to Ad

Re: [R] plot a list

Add this after the zm<- statement colnames(zm) <- sapply(z.l, colnames) On Mon, Sep 8, 2008 at 12:09 PM, stephen sefick <[EMAIL PROTECTED]> wrote: > I have one more question. The below example is revised to better > reflect the problem that I am running into. The two columns for each > data fr

[R] question on extracting binary data from pdf image

Dear List, If I use the following code to generate a pdf, pdf(filename) image(x, y, !is.na(z), col=c("green","black")) dev.off() Can someone extract the 0-1 data (that is, is.na(z)) from the Adobe source? Thank you. Xiaochun [[alternative HTML version deleted]] _

Re: [R] Problem with starting and using R

Hi Brian and Duncan, Many thanks for your responses. Setting the 'Current format' in 'Regional and language options' under Control Panel to English (Singapore) solved the R usage problem for me. I will have a go at R-patched build 46507 after it is released. Regards, Thomas On Mon, Sep 8, 2008

Re: [R] Question about multiple regression

I would test the speed before making such as assumption. Note that lm.fit is faster than lm and if they have the same x matrix then you can do many in one call by having y be a matrix. On Mon, Sep 8, 2008 at 12:05 PM, Dimitri Liakhovitski <[EMAIL PROTECTED]> wrote: > Dear R-list, > maybe some of

Re: [R] plot a list

I have one more question. The below example is revised to better reflect the problem that I am running into. The two columns for each data frame in the list are named the same because they are subsets of the same site, which has the same name in the larger data set. So when xyplot plots the merg

[R] Question about multiple regression

Dear R-list, maybe some of you could point me in the right direction: Are you aware of any FREE Fortran or Java libraries/actual pieces of code that are VERY efficient (time-wise) in running the regular linear least-squares multiple regression? More specifically, I have to run small regression mod

Re: [R] How to preserve date format while aggregating

As requested in the last line to every message to r-help a cutdown version of the data should be posted with the question. Suppose such a cut down version is this: DF <- data.frame(Date = ISOdatetime(2008, 1, c(1, 2, 5, 3, 4), 0, 0, 0), Subject = c(1, 1, 2, 2, 2)) # Then sort the data an

Re: [R] problem with Hmisc

Bob Flagg wrote: Dear All, I'm reading Frank Harrell's wonderful Regression Modeling Strategies book and ran into a problem following the example in Chapter 8. I'm working on platform: Ubuntu 8.04 (i486-pc-linux-gnu) R version: 2.7.2 (2008-08-25) and my command sequence was: library

Re: [R] (no subject)

I apologize, I forgot to type the title. On Mon, Sep 8, 2008 at 11:39 AM, Jason Thibodeau <[EMAIL PROTECTED]>wrote: > Hello all, > > I have a very large file (280k lines) containing three comma separated > variables. The first variable is a 0 or 1 depicting a pass or fail. The > other two are X a

[R] (no subject)

Hello all, I have a very large file (280k lines) containing three comma separated variables. The first variable is a 0 or 1 depicting a pass or fail. The other two are X and Y coordinates. Is there a good way I can represent this data in a chart/plot form other than using a 3d histogram? If I need

[R] Vorticity and Divergence

Hi all, I have some wind data (U and V components) and I would like to compute Vorticity and Divergence of these fields. Is there any R function that can easily do that? Thanks in advance for any help Igor Oliveira CSAG, Dept. Environmental & Geographical Science, University of Cape Town,

Re: [R] another XML package question

Duncan Temple Lang schrieb: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Antje wrote: Hi Duncan, thanks a lot for your explanations. I tried the following now to understand a bit more: data <- getNodeSet(doc, "//Data") xmlName(data[[1]]) xmlName(xmlRoot(data[[1]])) xpathApply(data[[1]], "

Re: [R] another XML package question

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Antje wrote: > Hi Duncan, > > thanks a lot for your explanations. > > I tried the following now to understand a bit more: > > data <- getNodeSet(doc, "//Data") > xmlName(data[[1]]) > xmlName(xmlRoot(data[[1]])) > xpathApply(data[[1]], "./*", xmlNa

Re: [R] Saving functions

On Monday, 08 September 2008, 15:46 (UTC+0100), Williams, Robin wrote: > I have written a function in a text editor and saved it with an .R > extension. It is saved in my working directory. How can I run it, do I > need to use source? Just to add to Jim's reply: if you want to have this funct

Re: [R] Saving functions

You can 'source' it in to define it and then call it as usual. No particular directory since you have control of that with the 'source' function. File extension is by convention and anyone will work since source is just going to read in an text file and execute it. On Mon, Sep 8, 2008 at 10:46 A

[R] Saving functions

Hi, Appologies for the simple nature of this question, I am unable to find the answer in manuals (EG and introduciton to R). I have written a function in a text editor and saved it with an .R extension. It is saved in my working directory. How can I run it, do I need to use source? If so, how

Re: [R] free workspace

On 08/09/2008 7:45 AM, GAF wrote: thanx for the quick reply! so this is what I get when I type in conflicts() [1] "part" "buzz" "distance" "huid" "in_whistle" "interruptions""match" [8] "overlap" "part" "part

  1   2   >