Re: [R] ROracle issue with shared library and RMySQL

R CMD ldd /usr/local/lib/R/library/ROracle/libs/ROracle.so should help. (Assuming this is Linux, a piece of information I am guessing from the kernel number. What or who is `Gentoo'?) You are looking for either unsatisfied dependencies or a lack of expected Oracle dependencies. On Sun, 23

Re: [R] kruskal wallis for manova?

> Hello, > Is there like the kruskal wallis test in relation to ANOVA (no > restrictions on normallity and variance homogenity) something (in R) > for MANOVA? You may know Fortunato Pesarin's book @book{multivaria:2001, key = {386}, author= {Fortunato Pesarin}, title = {Mu

[R] link between arima and arma fit

Hi dear sirs, I am wondering why the fit of the time serie x with an arima and the fit of diff(x) with an arma (same coeff p & d) differ one from another here are the output of R: %%% > modelarma<-arma(diff(x),c(7,5)) > modelarma Call: arma(x = diff(x

[R] merging variable and character

Hi R lovers! I'd like to know the trick to get a plot with a title associating variable and text I'd like to plot something like for (i in 1:10) plot(x[,i],ylab="Serie" & i) to get Serie 1 Serie 2 so on and so forth... of course it doesn't work. If somebody could give me an idea, it wou

Re: [R] merging variable and character

Hello Try ?paste plot(x[,i],ylab=paste("Serie", i)) Ivar Herfindal On Mon, 24 Nov 2003 10:51:11 +0100, STOLIAROFF VINCENT <[EMAIL PROTECTED]> wrote: Hi R lovers! I'd like to know the trick to get a plot with a title associating variable and text I'd like to plot something like for (i in 1

R: [R] merging variable and character

See ?paste plot(x[,i], ylab=paste("series",i,sep=" ")) does what you want. I think it should be a good idea to read (at least) some of several official manuals, contributed doc, books on R. best, vito - Original Message - From: STOLIAROFF VINCENT <[EMAIL PROTECTED]> To: <[EMAIL PROTEC

Re: [R] dvips function gives "Documents not found" error

This is a question about Hmisc, I believe, although you certainly did not mention it. Looking at the source code I see sys(paste("cd", tempdir(), sc, optionsCmds("latex"), tmp)) do my guess is that it is writing .dvi files in tempdir(). If you look up help on tempdir() you will see how to a

[R] median() for ordered factor {was "what does this mean .."}

> "PD" == Peter Dalgaard <[EMAIL PROTECTED]> > on 21 Nov 2003 15:08:09 +0100 writes: PD> "Liaw, Andy" <[EMAIL PROTECTED]> writes: >> > From: Peter Dalgaard [mailto:[EMAIL PROTECTED] >> > >> > John Christie <[EMAIL PROTECTED]> writes: >> > >> > > what does this

[R] statistical prediction for glm()

hello, I apologize that I poste this question again but I did not receive any replies on this topic and would like to know if the mail has been read over or if there is no method of statistical prediction computation for poisson errors in glm(). I am looking for a way to compute the model pr

[R] Warning calling par

Hi. I have a function that works well under R 1.7.1 But under 1.8.0 and 1.8.1 I get:Warning message: calling par(new=) with no plot And it doesnt plot The function is: mplotmeth1<-function(data1,data2,alpha,nr){ psfile<-paste("Meth1_",nr[1],"-",nr[length(nr)],".ps",sep="") posts

[R] Bollinger Bands

Is there a way to create Bollinger Bands without having to loop on the observations of a time serie? Any help appreciated Thanks __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help

Re: [R] Bollinger Bands

On Mon, 2003-11-24 at 09:19, [EMAIL PROTECTED] wrote: > Is there a way to create Bollinger Bands without having to loop on the > observations of a time serie? > > Any help appreciated > > Thanks Have you considered matlines() or matplot()? Each will generate a plot or add lines to an existing

[R] Questions on Random Forest

Hi, everyone, I am a newbie on R. Now I want to do image pixel classification by random forest. But I has not a clear understanding on random forest. Here is some question: As for an image, for example its size is 512x512 and has only one variable -- gray level. The histogram of the image look

[R] statistical prediction for glm()

The help page for "predict" says it is a generic function. Then command 'methods("predict")' reveals a function called "predict.glm", which has its own help page. Does that provide what you want? Also, I didn't find your earlier post in "www.r-project.org" -> "search" -> "R site se

Re: [R] dvips function gives "Documents not found" error

On Mon, 24 Nov 2003 10:53:48 + (GMT) Prof Brian Ripley <[EMAIL PROTECTED]> wrote: > This is a question about Hmisc, I believe, although you certainly did > not mention it. Looking at the source code I see > > sys(paste("cd", tempdir(), sc, optionsCmds("latex"), tmp)) > > do my guess is

Re: [R] statistical prediction for glm()

Martin - I can't figure out what question you are asking. Does either predict.glm() in the base package or cv.glm() in the boot package do what you want ? The theory for all of this is given in the references listed in the help pages for "glm" and "cv.glm". - tom blackwell - u michigan

Re: [R] statistical prediction for glm()

Martin - Do you want a goodness of fit measure for a Poisson glm ? I would start py plotting the residuals and doing an analysis of deviance on nested models. See McCullagh and Nelder, as well as more recent references. - tom blackwell - u michigan medical school - ann arbor - On Fri, 2

RE: [R] Bollinger Bands

I assume you are looking to apply some function, f, over the last k points of your time series for each point in the series (except for the first k-1 points). Let z be a vector that represents your time series. Then the j-th column of zl <- matrix(c(z,NA),length(z),k)[-seq(k-1),] # ignor

[R] Model checking for glms

Hello, I would like to have some advises concerning checking glms quality for prediction. I've a glm with Gamma family and link=log, and I would like to compute a Press statistic to check the predictivity of my model. Since my link function is not "identity", I wonder which "Hat" matrix I should t

[R] mle in the gamma model

Dear [R]-list, I'm looking for a classic equivalent of the wle.gamma function (library wle) that estimate robustly the shape and the scale parameters of gamma data. I have a vector of iid gamma rv : >data=rgamma(100,shape=10,scale=3) and a vector of their weights: >weights=c(rep(.5/70,70),rep(.2

goodness of fit for poisson glm - was:Re: [R] statistical prediction for glm()

I have to apologize, my subject was misleading or wrong. thanks for pointing me to the correct name/description. A goodness of fit measure for a Poisson GLM is what I am looking for. I am looking for a method similar to cv.glm(),to receive the estimate of prediction error. I will have a look

Re: [R] dvips function gives "Documents not found" error

Thanks to you both for your suggestions, and great to hear from the author himself! I may be thick, but I'm still having the problem. I did look at help for tempdir(), but tempdir() just *returns* the directory that's used; it doesn't allow me to change it. As a slightly different tack, I trie

[R] Is R multi-threaded?

Hi, I couldn't find a good answer for this anywhere in the docs, and I've looked through the code for 1.8.0; could someone who knows please confirm that R (for unix) is *not* currently multi-threaded? (I'm having trouble dynamically loading a C++ shared lib, and I want to make sure it's not thre

[R] RMySQL valid field names

I'm having some problems with valid field names when using RMySQL to interface R (version 1.7.0, under RedHat9.0), to MySQL (4.1.0-alpha). I think I've spotted the problem and a solution (which is working for me), but I wanted to share this with you as I may be missing something. (Note: I'm awar

Re: [R] Is R multi-threaded?

R for Unix is not of itself multi-threaded, but it can call libraries which are (notably a multi-threaded BLAS). R for Windows is multi-threaded. On Mon, 24 Nov 2003 [EMAIL PROTECTED] wrote: > I couldn't find a good answer for this anywhere in the docs, and I've > looked through the code for 1.

Re: [R] mle in the gamma model

library(MASS) ?fitdistr You can also use survreg (survival) and glm (with some help from functions in MASS for the shape). On Mon, 24 Nov 2003, Dominique Couturier wrote: > I'm looking for a classic equivalent of the wle.gamma function (library > wle) that estimate robustly the shape and the s

[R] How to get the parameters of nls(...) for later use

Hi all, I need to use the parameter estimates of nls() for further analysis. I know how to do in S+, e.g. nls(...)$parameters. In R, the attributes(nls(...)) does not have parameters, how would one get the parameter values out of nls()? Thanks in advance. Nancy __

[R] 1.8.1 and subsetting dataframes

I've encountered something that didn't arise using earlier versions of R (Linux). A dataframe is created and new columns added to it by doing calculations using apply with various functions on some of the original columns. It's somewhat too involved to give a toy example that's reproducible. How

Re: [R] How to get the parameters of nls(...) for later use

"Nancy Lo" <[EMAIL PROTECTED]> writes: > Hi all, > > I need to use the parameter estimates of nls() for further analysis. I > know how to do in S+, e.g. nls(...)$parameters. In R, the > attributes(nls(...)) does not have parameters, how would one get the > parameter values out of nls()? > Thanks

Re: [R] 1.8.1 and subsetting dataframes

The changes you point to were between 1.7.1 and 1.8.0, not between 1.8.0 and 1.8.1. 1.7.1 was quite capable of producing invalid data frames from erroneous usages. I think we really do need to see a reproducible example. On Tue, 25 Nov 2003, Patrick Connolly wrote: > I've encountered somethin

Re: [R] How to get the parameters of nls(...) for later use

"Nancy Lo" <[EMAIL PROTECTED]> writes: > I need to use the parameter estimates of nls() for further analysis. I > know how to do in S+, e.g. nls(...)$parameters. In R, the > attributes(nls(...)) does not have parameters, how would one get the > parameter values out of nls()? In both R and S-PLUS

Re: [R] 1.8.1 and subsetting dataframes

Patrick Connolly <[EMAIL PROTECTED]> writes: > I've encountered something that didn't arise using earlier versions of > R (Linux). > > A dataframe is created and new columns added to it by doing > calculations using apply with various functions on some of the > original columns. It's somewhat to

Re: [R] 1.8.1 and subsetting dataframes

On Mon, 24-Nov-2003 at 11:36PM +0100, Peter Dalgaard wrote: |> Patrick Connolly <[EMAIL PROTECTED]> writes: |> |> > I've encountered something that didn't arise using earlier versions of |> > R (Linux). |> > |> > A dataframe is created and new columns added to it by doing |> > calculations using

[R] Variance Ratios

Hello, I wrote a function to calculate the variance ratios like Lo and MacKinlay (1988): > vari <- function(indice,n,q){ + y <- rep(0,q) + indice.reg <- lm(indice[2:n]~-1+indice[1:(n-1)]) + a <- as.vector(acf(indice,q,plot=F)$acf) + b <- 1+2*( + for(k in 1:(q-1)){ + y[k+1]<-(y[k]+(1-k/q)*a[k+1])

RE: [R] kruskal wallis for manova?

Hi, you may also be interested in Marti Anderson's NPMANOVA program, described at http://www.stat.auckland.ac.nz/~mja/Programs.htm Cheers, Duncan * Dr. Duncan Mackay School of Biological Sciences Flinders University GPO Box 2100 Adelaide S.A.5001 AUSTRA

Re: [R] dvips function gives "Documents not found" error

On Mon, 24 Nov 2003 11:09:17 -0800, David Kelly <[EMAIL PROTECTED]> wrote: >So: Is there someway I can change the value of tempdir() so I can >influence Hmisc? Not after R has started, but if you set the TMP or TEMP environment variable to the name of an existing directory, that will be used

[R] Time series indexing/subsetting

R-listers: I may be asking too much from R, but is there a way to use time indexing on a time series object. For instance: > tsobject <- ts(1:12, start =1999, freq = 4) > tsobject Qtr1 Qtr2 Qtr3 Qtr4 19991234 20005678 20019 10 11 12 > tsobject[1999

Re: [R] Time series indexing/subsetting

[EMAIL PROTECTED] wrote: R-listers: I may be asking too much from R, but is there a way to use time indexing on a time series object. For instance: tsobject <- ts(1:12, start =1999, freq = 4) tsobject Qtr1 Qtr2 Qtr3 Qtr4 19991234 20005678 20019 10

[R] Y axis scale in plot.gam

Hi, Is there any way to change the y axis range of values in a plot.gam()? I need that two different GAM plots to be of the same scale. Also, it is possible to change the labels? I tried with "ylab" and "ylim" and did not work Thanks in advance Ricardo Lopes Ricardo Lopes ...

[R] Re: [RMySQL] unable to establish connection since R-1.8.0 upgrade

Hi Laurent, The issue is a very simple one to correct --- you only need to re-install RMySQL with the R version you're currently using. The issue is that the internal representation of objects in the methods package changed in 1.8.0, and thus packages that use S4 classes and methods and save thei