similar to: multi-level modeling & R?

Hello, everyone, I want to ask a simple question. If I have a set of data,and I want to identify how many outliers there are in the data.Which packages and functions can I use? Thanks. Shao chunxuan. [[alternative HTML version deleted]]

I am working on a function that will remove outliers for regression analysis. I am stating that a data point is an outlier if its studentized residual is above or below 3 and -3, respectively. The code below is what i have thus far for the function x = c(1:20) y = c(1,3,4,2,5,6,18,8,10,8,11,13,14,14,15,85,17,19,19,20) data1 = data.frame(x,y) rm.outliers =

I have been learning about some outlier tests -- Dixon and Grubb, specifically -- for small data sets. When I try help.start() and search for outlier tests, the only response I manage to find is the Bonferroni test avaiable from the CAR package... are there any other packages the offer outlier tests? Are the Dixon and Grubb tests "good" for small samples or are others more

Hi, I wonder if anybody could help me in converting this easy SAS program into R. (I'm still trying to do that!) PROC IMPORT OUT= WORK.CHLA_italian DATAFILE= "C:\Documents and Settings\carleal\My Documents\REBECCA\stat\sas\All&nutrients.xls" DBMS=EXCEL2000 REPLACE; GETNAMES=YES; RUN; data chla_italian; set chla_italian;

Hi there, I hope I can get advice regarding the calculation of leverage values or studentized residual values of a non-linear regression model. It seems like rstudent() does not work on a nls object. Many thanks in advance! Best regards, Xingli

Hello, I am learning to use the metafor package to conduct meta-regression analyses for a systematic review on multidisciplinary care interventions in chronic kidney disease. For the forest plots, I can't figure out how to plot unadjusted and adjusted models on the same plot. From top to bottom, I would like to be able have the unadjusted plot, the multivariate adjusted plot, then each

I have built rsync on IA64 using Intel's 'icc' compiler and can run the rsync executable interactively but when attempting to run under 'crontab' I get the following error: + /home/horace/mccssmb2/src/rsync-2.6.9/rsync -z --partial -v --progress --recursive --stats --times --links --exclude-from=/project/horace/mcc101/.rsync-EXCLUDE --timeout=1200 /project/horace/mcc101

Dear list, I know it must be obvious and I did my homework. (In fact I've RSiteSearched with keyword "remove AND library" but got timed out.(why?)....) How do I unload a library? I don't mean getting ride of it permanently but just to unload it for the time being. A related problem : I have some libraries loaded at startup in .First() which I have in .Rprofile. Now, I exited R

I am using the Unix version of R (version 1.7.0), installed via fink on a G4 Macintosh. I recently upgraded from version 1.6.0 and found that the "car" library now has a problem: ---Begin transcript--- >library(car) Attaching package 'car': The following object(s) are masked from package:base : dfbeta dfbeta.lm dfbetas dfbetas.lm hatvalues hatvalues.lm

Dear list, I know this is really basic question but I just couldn't get anything to work. (I did a R site search with keywords "zoo" and "data frame" but the server timed out on me.) I have a time series which has the following (typical) format, DATE Open High Low Close Volume 01-JAN-2006 5.25 5.25

Hi, I am trying to figure out how to get the correlation coefficient for a QQ plot (residual plot). So to be more precise, I am creating the plot like this: qq.plot(rstudent(regrname), main = rformula, col=1) But want to also access (or compute) the correlation coefficient for that plot. Thanks, Tom [[alternative HTML version deleted]]

Hi folks, Is there a function to show the size of an R object? eg. in Kbytes? Couple months ago Bendix Carstensen posted this marvelous little function lls(), which shows all objects in the current workspace by mode, class and 'size'. This is a wonderful enhancement to the build-in ls() already and I now have it sourced in my Rprofile.site at startup. The only drawback is,

The Partek package (www.partek.com) allows only two selections for Multiple Test Correction: Bonferroni and Dunn-Sidak. Can anyone suggest why Partek implemented Dunn-Sidak and not the other methods that R has? Is there any particular advantage to the Dunn-Sidak method? R knows about these methods (in R 2.1.1): > p.adjust.methods [1] "holm" "hochberg" "hommel"

Hi list, I'm sure the explanation must be laughably simple to the experts out there, but I just could figure it out. I have a simple data frame that looks like, >head(da.off) DDate OffP 1 2005-01-01 41.23 2 2005-01-02 44.86 3 2005-01-03 44.86 4 2005-01-04 43.01 5 2005-01-05 45.47 6 2005-01-06 48.62 where the first column DDate currently is character, and OffP is numeric. I want

Folks, I've entered into an R programming territory I'm not very familiar with, thus this probably very elementary question concerning the mechanic of a function call. I want to know from within a function the name of the variables I pass down. The function makes use of the "..." to allow for multiple unknown arguments, myfun = function(...) { do something } In the body I

Matt, Here's a Bonferroni-corrected multiple one-sample t-test that I wrote some years ago. It took a while to get it into R, as na.omit doesn't seem to handle vectors and I had to write a quick kludge (na.remove). Another more general point was that I discovered that the help page for t.test gives the name "parameters" for the degrees of freedom, as in S. However, the name

Hi! I need to run a wilcoxon (Mann-whitly, in fact) test with bonferroni correction, as I am running 10 consecutive wilcoxon test not independent, and I know that bonferroni will partially correct for this problem, but I have no idea how to do it with R, I have been looking in the archive but couldn't understand how to do it. The format I am using at the moment is r4_o <- [1]

Folks, I wonder if anything could be said about the distribution of a random variate x, where x = N(0,1)/N(0,1) Obviously x is pathological because it could be 0/0. If we exclude this point, so the set is {x/(0/0)}, does x have a well defined distribution? or does it exist a distribution that approximates x. (The case could be generalized of course to N(mu1, sigma1)/N(mu2, sigma2) and one

Hi- Is anyone working on multiple comparisons of means or post-hoc tests (ie: Tukey, Bonferroni) for R? I saw in the winter 98' archives of the R mailing lists that these tests had not been implemented yet, I was just wondering if I could look foward to having them. ;-) I also looked through the contributed packages and didn't see anything that offered such tests. I guess I could check

Hi, I'm having no luck figuring out how to find Bonferroni simultaneous confidence intervals to obtain a family of estimates in R. Does anyone know how to do this? Thank you!