similar to: More precision problems in testing with Intel compilers

I compiled the 1.9.1 src.rpm with the standard gnu tools and it works. I tried compiling the 1.9.1 src.rpm with the Intel 8 C and FORTRAN compilers and it bombs out during the testing phase: comparing 'd-p-q-r-tests.Rout' to './d-p-q-r-tests.Rout.save' ...267c267 < df = 0.5[1] "Mean relative difference: 5.001647e-10" --- > df = 0.5[1] TRUE

signif(.) is a <primitive> function. Unfortunately, I couldn't even find WHERE in the source, signif(.) is defined. Here are the symptoms: xmin <- .Machine $ double.xmin signif(xmin,3) #--> NA umach <- unlist(.Machine)[paste("double.x", c("min","max"), sep='')] for(dig in 1:10) {cat("dig=",dig,": ");

The original report was that compiling with gcc2.95.2 gcc and the related g77 on Irix 6.,5.13 produced an R which wailed its tests with: >> On entry to DGEMM parameter number 1 had an illegal value To some extend this confused me, as the source only seems to mention DGEMM in lowercase. Eventually, after the debug code I added did nothing, I realized that the R which was build was not

How does one loop over factors? Perhaps this is a newbie question. I tried: > b [1] caseX caseY caseZ Levels: caseX caseY caseZ > length(b) [1] 3 > > for (i in b) { + print (b == i) ; + print (i); + } [1] FALSE FALSE FALSE [1] 1 [1] FALSE FALSE FALSE [1] 2 [1] FALSE FALSE FALSE [1] 3 > But that strangely doesn't work. I must protest the implications of the above. i , as

It appears that the maximum number of connections available in R is about 48. Can anyone tell me how to bump this number up? I've been perusing the source, but any info would speed things up. Is there a reason that it was set to such a low number? Thanks for any help. -Frank

Hi, I am building R off the source file "The latest release (2004-06-21): R-1.9.1.tgz <http://cran.stat.ucla.edu/src/base/R-1.9.1.tgz> " on RH Linux 7.3. After the build the "make check" failed with an error which I judge to be minor, but would like to hear from the experts in the field. Is it harmeless? Can it indicate some problem with my computer setup (not R)?

Anybody know where I can get the Pinheiro/Bates book? I can't find a bookstore w/ stock and the publisher says they don't know when they'll have it again. Thanks. -Frank

I'm running R-2.0.0 (yesterday's snapshot)in its own directory, and everything works great, except: > .Random.seed Error: Object ".Random.seed" not found Does 2.0.0 not use .Random.seed for saving, etc, like it says in the help page? Thanks for any help. -Frank

R-1.7.0 built on NetBSD 1.6, but the validation test suite failed: Machinetype: Intel Pentium III (600 MHz); NetBSD 1.6 (GENERIC) Remote gcc version: gcc (GCC) 3.2.2 Remote g++ version: g++ (GCC) 3.2.2 Configure environment: CC=gcc CXX=g++ LDFLAGS=-Wl,-rpath,/usr/local/lib make[5]: Entering directory `/local/build/R-1.7.0/src/library' >>> Building/Updating

I checked the help and the mailing list archives, but I can find no mention of a routine that calculates higher moments like skewness and kurtosis. Of course, these are easy enough to write myself, but I was thinking that they MUST be in here. Am I wrong? Thanks. -Frank

Hi R-helpers Has anyone had any experience doing CGI or Servlets or using an httpd server in R? Context: we are interested in running dynamic demographic calculations on the internet (life tables, and such) in response to form submissions. I might do simple cgi to get started, but loading in data each time seems like it would be quite, quite slow. I have looked at Rserv, but I am most

This is too bad (and an old bug at least since 0.49); it also breaks some code I have been using in S-plus: > tst <- function(...) ls(...) > tst()#--- should really give the same as ls() !! character(0) > ls() [1] "%in%" "As.integer" "Dir" [4] "Dir0" "Fn12"

Full_Name: Jerry W. Lewis Version: 2.6.1 OS: Windows XP Professional Submission from: (NULL) (24.147.191.250) pchisq(0,0,ncp=lambda) returns 0 instead of exp(-lambda/2) pchisq(x,0,ncp=lambda) returns NaN instead of exp(-lambda/2)*(1 + SUM_{r=0}^infty ((lambda/2)^r / r!) pchisq(x, df + 2r)) qchisq(.7,0,ncp=1) returns 1.712252 instead of 0.701297103 qchisq(exp(-1/2),0,ncp=1) returns 1.238938

If anyone has a few extra CPU cycles to spare, I'd appreciate it if you could verify a problem that I have encountered. Run the code below and tell me if it crashes your R before completion. library(lme4) data(bdf) dump<-sapply( 1:50000, function(i) { fm <- lme(langPOST ~ IQ.ver.cen + avg.IQ.ver.cen, data = bdf, random = ~ IQ.ver.cen | schoolNR); cat("

Hi, I have a setup of Windows NT machines and SUN Solaris (UNIX) machines. I have installed samba on one of the UNIX machine. I am able to connect to the UNIX machine using samba from any of my NT machines. I want only few of my NT machines should be able to connect to the UNIX machine (with samba). To do so, I have set following in my smb.conf file hosts allow = hostname1, hostname2, ... I

Dear List: I am trying to find the area between two ECDFs. I am examining the gap in performance between two groups, males and females on a student achievement test in math, which is a continuous metric. I start by creating a subset of the dataframe male<-subset(datafile, female="Male") female<-subset(datafile, female="Female") I then plot the two CDFs via

Hello I have a problem when fitting a mixed generalised linear model with the lmer-function in the Matrix package, version 0.98-7. I have a respons variable (sfox) that is 1 or 0, whether a roe deer fawn is killed or not by red fox. This is expected to be related to e.g. the density of red fox (roefoxratio) or other variables. In addition, we account for family effects by adding the mother

i want to plot values with frequency on a logarithmic x axis. similar to this example that i found in the web: http://www.usspeaker.com/jensen%20p15n-graph.gif I would like to convert long numbers to si prefix notation like in the example (200000 to 200k, 35000000 to 3.5 M) Of course i could create labels by hand, but i have many files so i need some automatic function. Has anyone done that

Full_Name: Chanseok Park Version: 1.8.1 OS: linux-gnu Submission from: (NULL) (130.127.112.183) digamma with any negative value does not give a right answer. It gives -1.797693e+308 for any negative arguments. For example, digamma(-1.1) gives -1.797693e+308. The right answer should be 10.15416 This bug can be easily fixed by using the following digamma identity. digamma(x) = digamma(1-x) -

Hi all, I have noticed incorrect parsing of very small hexadecimal numbers like "0x1.00000000d0000p-987". Such a hexadecimal representation can can be produced by sprintf() using the %a flag. The return value is incorrectly reported as 0 when coercing these numbers to double using as.double()/as.numeric(), as illustrated in the three examples below: