Re: [R] nls: different results if applied to normal or linearized data

On Thu, 6 Mar 2008, Wolfgang Waser wrote: Thanks for your comments! Yes. You are fitting by least-squares on two different scales: differences in y and differences in log(y) are not comparable. Both are correct solutions to different problems. Since we have no idea what 'x' and 'y' are, we

[R] Problem with PCA

Dear Ms/Mr, In a package, I want to use PCA function. The structure I used follow this page: http://www.statmethods.net/advstats/factor.html. fit<-principle(mydata, nfactors=9, rotation=TRUE) or: result<-PCA(mydata) But I don't known why R language in my computer noticed: "not fou

[R] kruskal wallis post hoc test in R

Hello, I need nemenyi-test or any other post-hoc test for kruskal-wallis, but I just can't find out how to implement this in R. My data set is nitrite concentrations in four different groups which I intend to compare. The kruskal-wallis-test showed significance, but I still don't know between whi

Re: [R] testing for significantly different slopes

>> What's the problem? > > The problem is that I would like to do a pair-wise comparison > between the > multiple slopes. For example with this model: > > lm1 <- lm (Sepal.Length ~ Species/Sepal.Width -1, data=iris) > > # truncated output from summary(lm1) > # just the slope terms > Speciessetos

[R] Principle component analysis function

Dear All, In a package, I want to use PCA function. The structure I used follow this page: http://www.statmethods.net/advstats/factor.html. fit<-principle(mydata, nfactors=9, rotation=TRUE) or: result<-PCA(mydata) But I don't known why R language in my computer noticed: "not

[R] Repeated measures using lme

Dear list, I am trying to do a repeated analysis using lme in R and a little bit unsure if I have set up the right statement. The problem is the IL6 (interleukin 6) was measured 5 times on each individual in each of 6 companies. The hypotheses are to see whether there is a relationship be

Re: [R] Asking, are simple effects different from 0

On 3/5/2008 3:19 PM, jebyrnes wrote: > Indeed, but are not each of the cell means also evaluations of the effect of > one factor at the specific level of another factor? Is this an issue of > "Tomato, tomahto". I don't think it is "tomato, tomahto". Say the grand mean is around 100 and the w

Re: [R] box-constrained

Thank you very much, that is exactly what I wanted to know. I will add the gradient. Do you have any reference about : "the objective function is additive rather than multiplicative, it has better numerical conditioning" ? I am just being curious :) Gustave 2008/3/5, Ravi Varadhan <[EMAIL PROT

[R] Problem training svm with columns having same values

What should I do if I need to train svm() with data having same value in number of columns. These must be the deterministic features for the class. Is there anyway to over come this trouble? Regards - [[alternative HTML version deleted]]

Re: [R] Excel export into R

> I have this in excel > > Control > 543_BU > 123_AT > 432_CU > > > I want to be able to import to R so that it will read like this > > c<-c("543_BU","123_AT","432_CU") See the help page for read.csv ?read.csv ...and the R Data Import/Export manual http://cran.r-project.org/doc/manuals/R-data

Re: [R] Clustering large data matrix

Hi there, whether clara is a proper way of clustering depends strongly on what your data are and particularly what interpretation or use you want for your clustering. You may do better with a hierarchical method after having defined a proper distance (however this would rather go into statisti

[R] write in rimage package

Is there any way to write an imagematrix generated by the rimage package? Reproducible example: library(rimage) x <- read.jpeg(system.file("data", "cat.jpg", package="rimage")) png("newcat.png") plot(x) dev.off() The newcat.png image is different from the original file, because there are ugly w

Re: [R] histogram like x labels in barplot

If I understand: barplot(x) axis(1) On 06/03/2008, Hyunchul Kim <[EMAIL PROTECTED]> wrote: > Hi, all > > I drew a barplot with > > > barplot(data1, beside=TRUE) > > and then, I want to labels with hist() like x-axis ticks. > > How can I do this? > > Thanks in advance, > Hyunchul > >

[R] CREATE INTERFACE TO SELECT DIFERENT OPTIONS

Hello, I´m spanish student, and I´m making the finish project of computer science. I´m working in R and I need create a Interface which allow me select diferents execution options (similar to a menu). Is posible to create this menu,or interface, with R? or I have create this interface with other l

Re: [R] R_alloc with structures with "flexible array members"

Hi Dirk, Wow!!! That is really useful (and I was completely unaware of it). Thanks! ('sudo apt-get install littler' continued to work perfectly in all the machines I use). Thanks, R. P.D. One more reason to sign up for your tutorial on useR 2008. On Thu, Mar 6, 2008 at 2:35 AM, Dirk Eddelbu

[R] switch with a single character

Can anyone explain the results from switch below when a single charachter "E" is entered? It seems to work with letter S or N... > switch("East",West=1,East=2) # Correct [1] 2 > switch("E",W=1,E=2) # Not so correct [1] 1 > switch("E",E=1,W=2) # Even stranger [1] "E" > switch("S",N=1,S=-1) # Corr

[R] kruskal wallis post hoc test in R

Hello, I need nemenyi-test or any other post-hoc test for kruskal-wallis, but I just can't find out how to implement this in R. My data set is nitrite concentrations in four different groups which I intend to compare. The kruskal-wallis-test showed significance, but I still don't know between whi

Re: [R] Clustering large data matrix

Hi Dani, If you are working with NMR data, which data pretreatment methods you are using? 13112 variables for NMR data sounds too lot, you should apply some data binning or peak picking methods for data reduction. Also you must consider multicollinearity problems related to spectroscopic data, the

[R] histogram like x labels in barplot

Hi, all I drew a barplot with > barplot(data1, beside=TRUE) and then, I want to labels with hist() like x-axis ticks. How can I do this? Thanks in advance, Hyunchul __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PL

Re: [R] Principle component analysis function

phthao05 wrote: > Dear All, > In a package, I want to use PCA function. The structure I used follow this > page: http://www.statmethods.net/advstats/factor.html. >fit<-principle(mydata, nfactors=9, rotation=TRUE) >or: >result<-PCA(mydata) > > But I don't known why R languag

Re: [R] Excel export into R

Keizer_71 wrote: > > I have this in excel > > Control > 543_BU > 123_AT > 432_CU > > I want to be able to import to R so that it will read like this > > c<-c("543_BU","123_AT","432_CU") > > output: > [1] "543_BU" "123_AT" "432_CU" > > This is just a short version. I have about 20 rows and

[R] Clustering large data matrix

Hello, I have a large data matrix (68x13112), each row corresponding to one observation (patients) and each column corresponding to the variables (points within an NMR spectrum). I would like to carry out some kind of clustering on these data to see how many clusters are there. I have tried the

Re: [R] kruskal wallis post hoc test in R

On Thursday 06 March 2008 10:18:18 am Melanie Müller wrote: MM> Hello, MM> I need nemenyi-test or any other post-hoc test for kruskal-wallis, but I MM> just can't find out how to implement this in R. MM> My data set is nitrite concentrations in four different groups which I MM> intend to compare. T

[R] glm.cv delta value?

Hi! When I use the glm.cv function I get a value called "delta" which is explained as the "raw cross-validation estimate of prediction error". I recently found a formula for that term in literature where it is defined as: alpha = 1 / N * sum over( yi - yi,pred,CV) Well it is somehow similar to

Re: [R] write in rimage package

Perhaps: png("newcat.png") par(mar=rep(0, 4)) plot(x) dev.off() On 06/03/2008, Alberto Monteiro <[EMAIL PROTECTED]> wrote: > Is there any way to write an imagematrix generated by the rimage package? > > Reproducible example: > > library(rimage) > x <- read.jpeg(system.file("data", "cat.jpg",

Re: [R] read.table

Georg Ehret wrote: > Dear R community, I encounter a problem reading data into a dataframe. See > attachment for the input. I use: > data<-read.table("test",fill=T,row.names=1) > > When you look at "data" you can see that some lines of my input were > broken... I can avoid this problem by sortin

[R] order the plots using lattice

Dear all, I am running something similar to this short example: library("lattice") library("geneplotter") X=matrix(cbind(1:100,100:1,c(1:50,50:1)),ncol=3) Y=matrix(cbind(1:100,1:100,1:100),ncol=3) names=1:3 dummy.df = data.frame(names = factor(names, levels = names), x = seq_along(names), y

Re: [R] nls: different results if applied to normal or linearized data

On Thursday 06 March 2008 (07:03:34), Prof Brian Ripley wrote: > The only thing you are adding to earlier replies is incorrect: > > fitting by least squares does not imply a normal distribution. > Thanks for the clarification, 'implies' is to strong. I should have written 'suggests' or 'is

Re: [R] CREATE INTERFACE TO SELECT DIFERENT OPTIONS

Henrique Dallazuanna wrote: > Try this: > > switch(menu(c("Normal", "Uniform")), A=rnorm(5), B=runif(5)) > > for more details see ?menu > It might be more to the point to look into the tcltk package (or one of the other GUI packages) and its demos. (Also look at James Wettenhall's website). -

Re: [R] Problem with PCA

On Wed, 2008-03-05 at 19:43 -0800, phuong thao wrote: > Dear Ms/Mr, > In a package, I want to use PCA function. The structure I used follow this > page: http://www.statmethods.net/advstats/factor.html. >fit<-principle(mydata, nfactors=9, rotation=TRUE) >or: >result<-PCA(mydata)

Re: [R] Excel export into R

Hi Richie, I apologize. I thought i deleted this thread since it is resolved by using scan() function. thank you again, Kei Richard Cotton wrote: > >> I have this in excel >> >> Control >> 543_BU >> 123_AT >> 432_CU >> >> >> I want to be able to import to R so that it will read like this >

[R] Problem training svm with columns having same values

What should I do if I need to train svm() with data having same value across all rows in number of columns. These must be the deterministic features of the class and we cant exclude these columns to build model. The error I am getting is: Error in predict.svm(ret, xhold) : Model is empty! Is t

Re: [R] R_alloc with structures with "flexible array members"

Humm... but I mistakenly thought I was assuming almost nothing. R. On Wed, Mar 5, 2008 at 10:38 PM, Jeffrey Horner <[EMAIL PROTECTED]> wrote: > Ramon Diaz-Uriarte wrote on 03/05/2008 03:00 PM: > > > Dear Prof. Ripley, > > > > Yes, of course! You are right. What a silly mistake on my part! I w

Re: [R] Principle component analysis

On 3/5/08, phthao05 <[EMAIL PROTECTED]> wrote: > 1) I don't know why PCA rotation function not run although I try many times. > Would you please hepl me and explain how to read the PCA map (both of > rotated and unrotated) in a concrete example. If you used the example from here [1], there's a

[R] Interface or Select menu

Hello, I´m spanish student, and I´m making the finish project of computer science. I´m working in R and I need create a Interface which allow me select diferents execution options (similar to a menu). Is posible to create this menu,or interface, with R? or I have create this interface with othe

[R] replace NA with 9999 in zoo object

This is the same set of data that I have been working with for those in the know. it is a matrix of ~174 columns and ~70,000 rows. I have it as a zoo object, but I could read it in as just a matrix as long as the date time stamp won't be corrupted. here is an example of what a column would look

[R] Try to save as PDF: font family not found in PostScript font database

Hi I'm using 2.6.0 on Windows XP SP2. I realise I'm on an old version, please tell me if the problem is fixed in the newer versions. When I try and use the graphics window menu to save a graph as PDF, I get the following error, and no PDF produced Error: Invalid font type In addition: Warning m

Re: [R] CREATE INTERFACE TO SELECT DIFERENT OPTIONS

Try this: switch(menu(c("Normal", "Uniform")), A=rnorm(5), B=runif(5)) for more details see ?menu On 06/03/2008, ermimi <[EMAIL PROTECTED]> wrote: > > Hello, I´m spanish student, and I´m making the finish project of computer > science. I´m working in R and I need create a Interface which allow

Re: [R] Principle component analysis function

phthao05 wrote: > Dear All, > In a package, I want to use PCA function. The structure I used follow this > page: http://www.statmethods.net/advstats/factor.html. >fit<-principle(mydata, nfactors=9, rotation=TRUE) >or: >result<-PCA(mydata) > > But I don't known why R languag

Re: [R] Interface or Select menu

er MIMI & piki PIKINHA wrote: > > Hello, I´m spanish student, and I´m making the finish project of > computer science. I´m working in R and I need create a Interface > which allow me select diferents execution options (similar to a menu) > . Is posible to create this menu,or interface, with R? o

[R] Odp: replace NA with 9999 in zoo object

Hi not sure about zoo but data frame or matrix can be indexed mat[is.na(mat)] <- Regards Petr [EMAIL PROTECTED] [EMAIL PROTECTED] napsal dne 06.03.2008 14:09:16: > This is the same set of data that I have been working with for those > in the know. it is a matrix of ~174 columns and ~70,0

Re: [R] loop

Hi [EMAIL PROTECTED] napsal dne 05.03.2008 18:04:35: > Thank you Yinghai, that's what I need :-)! > > Yinghai Deng <[EMAIL PROTECTED]> schrieb: m <- seq(-1,1,0.1) > x1 <- c() > x2 <- c() > for(i in 1:length(m)){ > x1[i] <- m[i] > x2[i] <- m[i]^2 > } > dat <- data.frame(x1,x2) The

Re: [R] differentiating a numeric vector

Thanks for all the ideas. splinefun looks like the simplest way to achieve what I need: > x <- 1:10 > y <- x^2 > f <- splinefun(x,y) > f(3,deriv=0) [1] 9 > f(3,deriv=1) [1] 6 > f(3,2) [1] 2 > f(3,3) [1] -3.330669e-16 The fda package has a function bsplineS which does nearly the same thing, and t

Re: [R] replace NA with 9999 in zoo object

It works just the same as matrices: > z <- zoo(cbind(a = c(1, NA, 3), b = c(NA, 10, 11))) > z[is.na(z)] <- 999 > z a b 1 1 999 2 999 10 3 3 11 > There are also a number of other methods for handling NAs in zoo: na.approx na.contiguous na.locf na.spline na.trim and na.stinterp in the

[R] Statistical Questions: finding differentially expressed genes

Hi Everyone, I am trying to find a way to do this in excel to tell me which genes are the most differentially expressed. Sorry, i couldn't find excel forum section in nabble. However, if it is in R it is fine. This is a microarray data, and it has been normalized. According to Dov Stekel in Micro

[R] 1-pnorm values in a table

Hi, I've read in a csv file (test.csv) which gives me the following table: Hin1 Hin2 Hin3Hin4 Hin5 Hin6 HAI1 9534.83 4001.74 157.16 3736.93 484.60 59.25 HAI2 13272.48 1519.88 36.35 33.64 46.68 82.11 HAI3 12587.71 5686.94 656.62 572.29 351.60 136.91 HAI4

Re: [R] 1-pnorm values in a table

> I've read in a csv file (test.csv) which gives me the following table: > >Hin1 Hin2 Hin3Hin4 Hin5 Hin6 > HAI1 9534.83 4001.74 157.16 3736.93 484.60 59.25 > HAI2 13272.48 1519.88 36.35 33.64 46.68 82.11 > HAI3 12587.71 5686.94 656.62 572.29 351.60 136.91 > H

Re: [R] Try to save as PDF: font family not found in PostScript font database

Hi Michael, I had the same problem using R 2.6.0. Not it's fixed in newer versions. HTH Jorge On Thu, Mar 6, 2008 at 8:05 AM, michael watson (IAH-C) < [EMAIL PROTECTED]> wrote: > Hi > > I'm using 2.6.0 on Windows XP SP2. I realise I'm on an old version, > please tell me if the problem is fix

Re: [R] switch with a single character

On Thu, 6 Mar 2008, Henrik Andersson wrote: > Can anyone explain the results from switch below when a single > charachter "E" is entered? It seems to work with letter S or N... ?switch . . . Warning: Beware of partial matching: an alternative 'E = foo' will match the first argument

Re: [R] 1-pnorm values in a table

This is a bit ugly but I think it works. myf <- function(x) 1-pnorm(x,mean(x), sd(x)) results <- apply(test, 2, myf) mymeans <- apply(test, 2, mean); mymeans for (i in 1:length(test)){ test[,i][test[,1]>=mymeans[i]] <- NA } results[is.na(tes

Re: [R] 1-pnorm values in a table

hsl.gov.uk> writes: [snip] > Try: > nrows <- 5 > mm <- matrix(rnorm(30),nrow=nrows) > sd.by.col <- apply(mm,2,sd) > mean.by.col <- apply(mm,2,mean) > values <- 1-mapply(pnorm, q=as.vector(mm), mean=rep(mean.by.col, nrows)), > sd=rep(sd.by.col, nrows))) > values <- matrix(values, nrow=5) > >

Re: [R] Softmax in nnet

(this is a resend) Hi R help, I run my data in nnet with skip layer, factor response (with 0 & 1 values) and explicitly put softmax=T to compare the result of the default nnet with no softmax specification. I assume this should give me the same result. I got the result from the default one, b

Re: [R] Asking, are simple effects different from 0

Ah, I see where we are talking past each other. In my particular analysis (I'm looking at deviations from a predicted value), any deviation from 0 (whether due to grand mean or not) is actually very very interesting. What is ultimately interesting to me is the sign of that difference, but, I nee

Re: [R] ipf function in R

Chandra, While I can't advise on how to use the ipf function in the cat package, I can offer the following function that we use here to balance rebalance arrays with new marginal totals. #ipf.R #Function to iteratively proportionally fit a multidimensional array #IPF also known as Fratar method,

Re: [R] 1-pnorm values in a table

hsl.gov.uk> writes: [snip] > Try: > nrows <- 5 > mm <- matrix(rnorm(30),nrow=nrows) > sd.by.col <- apply(mm,2,sd) > mean.by.col <- apply(mm,2,mean) > values <- 1-mapply(pnorm, q=as.vector(mm), mean=rep(mean.by.col, > nrows)), sd=rep(sd.by.col, nrows))) values <- matrix(value

Re: [R] Variable importance in Random Forests

Depending on the way you ran randomForest, the variable of interest may have been used in many splits in many trees. The information can be extracted using the getTree() function. However, I suspect it might be more useful to look at partialPlot(). Best, Andy > -Original Message- > Fro

Re: [R] legend for several graphics

I think you need a par(xpd=NA) before the legend command (At least it did not show up for me until I set xpd). For positioning the overall title and legend, one option is the cnvrt.coords function in the TeachingDemos package. To center the overall title over the central column plots (the defau

Re: [R] mean and sd with number of values?

Hi Martin, Try x=c(NA,NA,rnorm(50),NA) sum(!is.na(x)) HTH Jorge On Thu, Mar 6, 2008 at 11:41 AM, Martin Kaffanke <[EMAIL PROTECTED]> wrote: > Hi there, > > When i do > > mean(fl[1:20], na.rm=T) > sd(fl[1:20], na.rm=T) > > I get 20 Results, but now I'd like to know from how many numbers my me

[R] Odp: mean and sd with number of values?

Hi [EMAIL PROTECTED] napsal dne 06.03.2008 17:41:06: > Hi there, > > When i do > > mean(fl[1:20], na.rm=T) e.g. sum(!is.na(fl[1:20])) Regards Petr > sd(fl[1:20], na.rm=T) > > I get 20 Results, but now I'd like to know from how many numbers my mean > and sd are calculated. > > How to I u

Re: [R] legend for several graphics

On Thu, 2008-03-06 at 09:41 -0700, Greg Snow wrote: > I think you need a par(xpd=NA) before the legend command (At least it > did not show up for me until I set xpd). Hi Greg, Yes, you are correct. Apologies to the OP, Georg. I must have set xpd = NA during my tests and not reset it, so when I ra

[R] mean and sd with number of values?

Hi there, When i do mean(fl[1:20], na.rm=T) sd(fl[1:20], na.rm=T) I get 20 Results, but now I'd like to know from how many numbers my mean and sd are calculated. How to I use the length() for this, or is this possible? Because there are missing values in some, but not in all columns. So ther

Re: [R] mean and sd with number of values?

Martin - Does the following help? sapply(fl, function(x) sum(!is.na(x))) See ?sapply Best, Erik Iverson Martin Kaffanke wrote: > Hi there, > > When i do > > mean(fl[1:20], na.rm=T) > sd(fl[1:20], na.rm=T) > > I get 20 Results, but now I'd like to know from how many numbers my mean > and s

[R] Tcl/Tk not working on Mac OS X

Greetings. Yesterday I decided to try out Rcmdr, but it refused to load. I traced the problem to the library tcltk, which when I tried to load it told me this: > > library(tcltk) > Loading Tcl/Tk interface ... Error in fun(...) : > Can't find a usable init.tcl in the following directories:

Re: [R] Odp: mean and sd with number of values?

Am Donnerstag, den 06.03.2008, 17:47 +0100 schrieb Petr PIKAL: > Hi > > [EMAIL PROTECTED] napsal dne 06.03.2008 17:41:06: > > > Hi there, > > > > When i do > > > > mean(fl[1:20], na.rm=T) > > e.g. > > sum(!is.na(fl[1:20])) This seems to give me a sum of all the items. But I'd like to have

Re: [R] Tcl/Tk not working on Mac OS X

Hi Aaron: Look at: http://article.gmane.org/gmane.comp.lang.r.mac/2305 Do take time to read the R Posting FAQ. A query to the mail-list archive would have shown that this problem has been discussed repeatedly. HTH, -Roy M. On Mar 6, 2008, at 9:34 AM, Aaron Solomon Adelman wrote: > Greet

[R] calculate AUC and plot ROC in R

Hi, there: Could someone tell me a simple function of plot ROC curve and calculate AUC in R? My setting is very simple, a column of the true binary response and another column of predicted probabilities. Thanks! Yulei [[alternative HTML version deleted]] __

Re: [R] calculate AUC and plot ROC in R

He, Yulei wrote: > Hi, there: > > > > Could someone tell me a simple function of plot ROC curve and calculate > AUC in R? My setting is very simple, a column of the true binary > response and another column of predicted probabilities. > > > > Thanks! > > > > Yulei ROC area is easy. Se

Re: [R] calculate AUC and plot ROC in R

He, Yulei hcp.med.harvard.edu> writes: > Could someone tell me a simple function of plot ROC curve and calculate > AUC in R? My setting is very simple, a column of the true binary > response and another column of predicted probabilities. There are a few packages with examples around, which you c

[R] build options for R

Hello Everyone, Recently I was given a Java servlet based web calculator that will call R (libR.so to be exact) but I am having trouble trying to disable R from requiring --save, --no-save or --vanilla... Is there any way to build R on linux and disable R from asking --save, --no-save or --

[R] Array arithmetic

I have two arrays A and B with dimensions of (L, M, N, P) and (L, M, N), and I want to do for (i in 1:L) { for (j in 1:M) { for (k in 1:N) { if (abs(B[i, j, k]) > 10e-5) C[i, j, k,] <- A[i, j, k,]/B[i, j, k] else C[i, j, k,] <- 0 } } } How can I get C more efficiently than looping? Thanks

[R] R-Logo in \LaTeX

Dear everybody! Is there a command in \LaTeX to display the R-Logo or has anybody made it up? Thank you in advance. Yours, sincerely Mag. Ferri Leberl __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the po

Re: [R] Array arithmetic

apparently you forgot the "commented, minimal, self-contained, reproducible code" part... L = 10 M = 20 N = 30 P = 40 set.seed(1) A = array(rnorm(L*M*N*P), dim=c(L, M, N, P)) B = array(rnorm(L*M*N), dim=c(L, M, N)) B[sample(100, 10)] = 0 C = array(0, dim=c(L, M, N, P)) for (i in 1:L) { for (j i

Re: [R] Array arithmetic

I think this should work: array(A[abs(B) > 10e-5]/B[abs(B) > 10e-5], dim=c(L, M, N, P)) On 06/03/2008, Gang Chen <[EMAIL PROTECTED]> wrote: > I have two arrays A and B with dimensions of (L, M, N, P) and (L, M, > N), and I want to do > > for (i in 1:L) { > for (j in 1:M) { > for (k in 1:N) {

Re: [R] Array arithmetic

Hum you are rigth, I forgot of 'else'. On 06/03/2008, Benilton Carvalho <[EMAIL PROTECTED]> wrote: > no, it won't. > > you're doing the right math on the "valid" subset... but you're not > returning the zeros where needed therefore, the whole thing will > get recycled to match the dimension

[R] Help with parsing a data file

Hi All, I need to parse data from a file, example shown below. The first two lines can be skipped, the third line contains the column names. The next 13 lines can be skipped. The next line "1991" is a year value, with the following 13 values data for that year. The file then repeats this forma

Re: [R] Correlation matrix one side with significance

Am Mittwoch, den 05.03.2008, 14:38 -0300 schrieb Henrique Dallazuanna: > Try this: > > On 05/03/2008, Martin Kaffanke <[EMAIL PROTECTED]> wrote: > > Hi there! > > > > In my case, > > > > cor(d[1:20]) > > > > makes me a good correlation matrix. > > > > Now I'd like to have it one sided, means

Re: [R] Correlation matrix one side with significance

Please provide a example of what you want On 06/03/2008, Martin Kaffanke <[EMAIL PROTECTED]> wrote: > > Am Mittwoch, den 05.03.2008, 14:38 -0300 schrieb Henrique Dallazuanna: > > > Try this: > > > > On 05/03/2008, Martin Kaffanke <[EMAIL PROTECTED]> wrote: > > > Hi there! > > > > > > In my

Re: [R] Array arithmetic

no, it won't. you're doing the right math on the "valid" subset... but you're not returning the zeros where needed therefore, the whole thing will get recycled to match the dimensions. b On Mar 6, 2008, at 2:03 PM, Henrique Dallazuanna wrote: I think this should work: array(A[abs(B)

[R] Storing output in an array or matrix

I have a data set which has, let's say, income by state. I'm trying to output income quartiles for each state into an array by doing a loop so that I don't have to do it state by state by hand. ie, something like for (i in 1:50) { quantile(subset(data,state==i)$income) -> r[i,5] } where

Re: [R] Correlation matrix one side with significance

On 3/6/2008 2:07 PM, Martin Kaffanke wrote: > Am Mittwoch, den 05.03.2008, 14:38 -0300 schrieb Henrique Dallazuanna: >> Try this: >> >> On 05/03/2008, Martin Kaffanke <[EMAIL PROTECTED]> wrote: >>> Hi there! >>> >>> In my case, >>> >>> cor(d[1:20]) >>> >>> makes me a good correlation matrix. >>>

[R] Help with colinearity problem in multiple linear regression

Hello, For basic linear regression lm() does the job well, for datasets that are larger than memory biglm() seems to work. I'm working on a parallel implementation of multiple linear regression for datasets that are too large for memory. Currently I am working over least squares: calcu

Re: [R] Storing output in an array or matrix

I think you could get this done with the function "tapply", but it's not clear because you don't give an example of your data. If tapply is not the answer, certainly there is a way that is better than resorting to doing it 'by hand' (brute force). See ?tapply Best, Erik Iverson tieflingrogue

Re: [R] Help with parsing a data file

Sean I'm sure there are many ways of doing this. I assume you have read the data in R as a data.frame with 24 columns and 2+1+13+(1+13)*n rows, where n is the number of years, and that you want a data.frame with 25 columns (one extra for year) and 13*n rows (although I am not sure why 13 MOnths)

Re: [R] order the plots using lattice

On Thu, 06-Mar-2008 at 12:00PM +, audrey wrote: [...] |> In the output, the first plot is represented in the bottom left |> corner, I would like the plots to be drawn from the top left corner |> but I cannot find in the Lattice help, the argument to set to do IIRC, it's as.table = TRUE Look

Re: [R] Help with parsing a data file

Try this: lines <- readLines('yourfile') newLines <- lines[-(1:(13+3))] coln <- scan(textConnection(lines[3]), what="") lapply(which(nchar(newLines) == 4), function(x)read.table(textConnection(newLines[seq(x + 1, x + 13)]), col.names=coln)) On 06/03/2008, sean <[EMAIL PROTECTED]> wrote: > Hi Al

[R] Interesting remarks about R back in 1999

Hi, this is not an R-help post, but I found this extract below that was written by a leading mathematician back in 1999 when he was talking about statistics and computing. I found it interesting to share and I ask your opinion do you think this still holds today or things have changed? Thanks. “.

Re: [R] x-only zoom and pan?

Peter Wolf wiwi.uni-bielefeld.de> writes: > > Hallo here is a simple proposal using tcltk-sliders. > > Peter Wolf > > > # step 3: test it. > > ts.zoom(runif(1), rexp(1), rnorm(1)) > > Randy Zelick wrote: > > >Hello list, > > > >Could the following be done without too much grie

Re: [R] CREATE INTERFACE TO SELECT DIFERENT OPTIONS

ermimi wrote: > > Hello, I´m spanish student, and I´m making the finish project of computer > science. I´m working in R and I need create a Interface which allow me > select diferents execution options (similar to a menu). Is posible to > create this menu,or interface, with R? or I have create

[R] Changing code within a package

I was able to hack the code in the polspline package in order to change the plot.polymars function so that I can set the limits of my y-axis. However, I was only able to do this by entering plot.polymars in R, copying the code to Word, changing what I needed, and then pasting the new code in R to r

Re: [R] Array arithmetic

Thank both of you for the suggestions! Gang On Mar 6, 2008, at 2:12 PM, Henrique Dallazuanna wrote: > Hum you are rigth, I forgot of 'else'. > > On 06/03/2008, Benilton Carvalho <[EMAIL PROTECTED]> wrote: >> no, it won't. >> >> you're doing the right math on the "valid" subset... but you're no

[R] Rpart and bagging - how is it done?

Hi there. I was wondering if somebody knows how to perform a bagging procedure on a classification tree without running the classifier with weights. Let me first explain why I need this and then give some details of what I have found out so far. I am thinking about implementing the bagging proc

Re: [R] Changing code within a package

On 3/6/2008 3:57 PM, Shewcraft, Ryan wrote: > I was able to hack the code in the polspline package in order to change > the plot.polymars function so that I can set the limits of my y-axis. > However, I was only able to do this by entering plot.polymars in R, > copying the code to Word, changing wh

[R] can't merge zoo ojects and convert to ts (been trying for 2 days)

I'm stuck, but am sure it can be done I just don't understand how. I have data in an irregular timeseries. I want to be able to use stl to visualise the data (see seasonal parts etc), so I need to change to regular series of class ts (I think). I am using 2 zoo objects one is regular and the othe

[R] R code for selecting places spatially and by time

Hi all The code of trying to write relates to selecting properties (given by x and y co-ordinates) spatially (distance "X" from "infected" properties identified by date) over a certain time period. i.e. what properties are within 3 km from properties infected on "2008-01-01" over the last 14 day

[R] How to hold a value(Mean sq) with a string

Hi all: Can someone advice me on how to hold the residuals Mean sq value on a string so it can be used in other calculations. I was trying something like this: Msquare<-dfr$Mean sq but fails..Thanks dfr <- read.table(textConnection("percentQ Efficiency 1.5650.0125 1.94 0.0213 0.

Re: [R] How to hold a value(Mean sq) with a string

Assign your anova object a name, like fm1 <- anova(lm(Efficiency~percentQ,data=dfr)) Then do names(fm1) and try fm1$"Mean Sq" Notice you were trying to find "Mean Sq" in dfr, your original data.frame. -Erik Iverson Felipe Carrillo wrote: > Hi all: > Can someone advice me on how to hold th

Re: [R] How to hold a value(Mean sq) with a string

Hi Felipe, Try this dfr <- read.table( textConnection("percentQ Efficiency 1.5650.0125 1.94 0.0213 0.8760.003736 1.0270.006 1.5360.0148 1.5360.0162 2.6070.02 1.4560.0157 2.16 0.0103 1.6980.0196 1.64 0.0098684 1.8140.0183 2.3940.0107 2.4690.

[R] Sweave and extra line

Hi the list, I am using Sweave. Between the Sinput and the Soutput, there is always an extra line: > 2+2 [1] 4 Is it possible to remove it ? > 2+2 [1] 4 Thanks Christophe __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinf

[R] Sweave and the prompt

Hi the list I would also like to remove the prompt, in order to let the reader cut-and-paste from pdf to R. So I set the prompt to " " with options(prompt=" ",continue=" ") But it add an extrat space at the start of each line. I also try some code options(prompt="\a",continue="\a") After a big

Re: [R] colored 3d scatter plot

Thank you all for your help. SNN wrote: > > > > Hi All, > > I have data for two groups, group with 100 points and group B with 15 > points. i needed plot these two groups in one scatter plot, each group > with a different color. I tried > > plot3d(data, col = c("red", "blue")[c(rep(1, 100)

  1   2   >