similar to: changing degrees of freedom in summary.lm()

Dear all, I'm wanting to do a series of comparisons among 4 categorical variables: a <- aggregate(y, list(var1, var2, var3, var4), sum) This gets me a very nice 2-dimensional data frame with one column per variable, BUT, as help for aggregate says, <<empty subsets are removed>>. I don't see in help(aggregate) how I can change this. In contrast, a <- tapply(y,

Hello, all; I get an unexpected result when trying to plot a probability histogram with R1.9.1 on windows xp: #with the following code: > x <- runif(100,0,1) > hist(x) > hist(x, freq=F) > h <- hist(x, freq=F) > summary(h) # Length Class Mode #breaks 11 -none- numeric #counts 10 -none- numeric #intensities 10 -none- numeric #density 10

Hi all, what a great help list! I hope someone can help me with this puzzle... I'm trying to find a simple way to do: boxplot(obs~factor) so that the factors are ordered left-to-right along the x-axis by median, not alphabetically by factor name. Complicated ways abound, but I'm hoping for a magical one-liner that'll do the trick. Any suggestions would be treasured. Thanks,

Hi all; Thanks for the responses to my query of how to make tapply into a table instead of an n-dimensional array. Summary of responses follows: Peter Dalgaard: as.data.frame(with(tmp,as.table(tapply(C,list(A=A,B=B),sum)))) Phil Spector wrote: z = tapply(y,list(var1,var2,var3,var4),sum) data.frame(do.call('expand.grid',dimnames(z)),y=do.call('rbind',as.list(z))) Hans

Greetings, I'm trying to find a function to generate a random landscape consisting of x by y cells, allows a specification of number of classes and their proportional abundance in the landscape. Does R support such a function ? Thanks in advance Steve Steve Friedman, Assistant Professor Center for Global Change and Earth Observations Michigan State University East Lansing, Michigan

> -----Original Message----- > From: JLB [mailto:jlb@twu.net] > I'd go further and cut down on membership to people emailing from > non-Windows systems (including Webmail systems, until someone > invents a worm that spreads through those...) Okay, I'm sorry, but that's the dumbest idea yet. If they didn't have Windows, why would they use Samba?

JLB <jlb@twu.net> wrote: > I'll be a bit less polite and a bit more to the point. > > If Unix systems didn't insist upon scattering files all the hell over > /etc, /usr/local/bin, /usr/local/lib, /usr/local/include, > /var/share/mybunghole, /opt/local/stupidlib, etc. etc... Anyone interested in _why_ un*x systems use this structure, take a look at the FHS

Hi All. I'd like to change the default plotting colors used to construct a scatterplot with regression line in the CAR package. So, scatterplot(y~pred,smooth=FALSE, xlab="X", ylab="Y", lwd=2) If I change the palette (e.g., palette(ranbow(6)), I can change the color of the lines and points. However, the axes and labels remain in black (i.e., the first color in the

Hi All. I need a bit of help extracting the residual error variance from the VarCorr structure from lmer. #Here's a 2-way random effects model lmer.1 <- lmer(rating ~ (1|person)+(1|rater), data = dat) #Get the structure vc.fit <- VarCorr(lmer.1) #results in..... $person 1 x 1 Matrix of class "dpoMatrix" (Intercept) (Intercept) 0.7755392 $rater 1 x 1 Matrix

Hello R users- I am new to R, and tried searching the archives and literature for an answer to this - please be patient if I missed something obvious. I am fitting a logistic regression model, and would like to include variance functions (specifically the varIdent function). I cannot figure out how to do this either in glmmPQL (or something similar) for the model with random effects, or in glm

Hi R helpers, I was wondering if anybody knows if is possible to generate bootstrap values for a cluster analysis in R. What I am trying to do is obtain some confidence on the clusters formed by resampling the data set. A similar type of analysis is used in molecular taxonomy and the confidence values of each cluster are placed in the nodes of the dendogram. Any ideas on how to do this

There are still some seats available in the workshop on R that John Fox will be teaching next week in Berkeley, CA. The workshop title is "The R Statistical Computing Environment: The Basics and Beyond." It will run from Monday, August 6 through Thursday, August 9 (9:00 a.m. through 5:00 p.m. each day). This workshop is offered as part of the Inter-university Consortium for

Hello all, I am trying to use "clim.pact" package for my work, but since this is the beginning for me to use gridded datasets in "R", I am having some trouble. I want to do seasonal analyses like trends, anomalies, variograms, EOF and probably kriging too to downscale my 1 degree gridded data to 0.5.  So, as a first step, I compiled my entire dataset (with 25

I understand that Rcmdr needs the tcltk package in order to run. However, when I attempt to using the library (Rcmdr) command, I get the message below. I have attempted to start X in R as well as using starting it from the Utilities folder on my Mac to no avail. I can't figure out if there is an issue with X11 or with R itself (I just updated to 1Is there a terminal command I can use

regarding this old message: > > Am Sam, M?r 20, 2004 at 04:40:10 -0500 schrieb JLB: >> Nopers, and it FREAKS OUT when I go into Printer Setup and select a page >> orientation (e.g. Portrait) > > Hmmm, seems that you first have to learn how to configure printing? Look > in the Documentation section on winehq for that. If you use Cups > everything is done for you

Btw, all the pbinom() examples posted to S-news today return the proper result in R Martin >>>>> "JS" == James Stapleton <stapleton at stt.msu.edu> writes to S-news : JS> More interesting behavior of pbinom: >> pbinom(3,2,.8) JS> [1] 1 >> pbinom(30,2,.8) JS> [1] 0.9967653 >> pbinom(40,2,.8) JS> [1]

Hi everybody I was trying the R binary (1.7.0) for Macintosh (Carbon) and I tried to install some packages using "install.packages" and apparently this function was not implemented. I reloaded the base library since the "update.packages" function should be on it but did not work. Also I tried read.table and did not work either. Does anybody know if the base package

Hi All. I'm new to R and trying to learn the data manipulation routines. At the moment, this is the one that has me stumped. Imagine n test takers provide responses to k multiple choice items. The result is a (n+1) x k matrix of responses where the first row of the matrix is a correct answer key. How could you transform the raw item responses into 1s (correct response) and 0s (incorrect

Hello All. I would like to repeat a multi-screen, multivariate time series plot across multiple units. Consider the following data: <r-help@stat.math.ethz.ch> UNIT TRIAL Y1 Y2 Y3 Y4 Y5 Y6 1 109 1 NA NA NA 13718 3655 15240 2 109 2 15000 6000 30000 13057 6198 21962 3 109 3 10000 10000 20000 8356 12929 27992 4 109

Hi, I would like to run R2.5.1 in Ubuntu feisty on an AMD64 chip. I am a total newbie to Linux. I have successfully compiled 2.5 however, I am not able to figure out how to upgrade the packages. I am not able to find an R-base-core_2.5.1-1_amd64 or all on the CRAN mirrors. I can find one on the Ubuntu feisty site, however when I try to install it I get a warning that Dependency is not stisfiable: