[R] Check errors

[Christian Hennig]

Dear group,

On 21 Jul 2000, Peter Dalgaard BSA wrote:

> Christian Hennig <fm3a004 at math.uni-hamburg.de> writes:
> 

(I ran "make check" to get:)

> > running code in base-Ex.R ...
> > ./../bin/R --vanilla < base-Ex.R > base-Ex.Rout
> > make[4]: *** [base-Ex.Rout] Error 1
> 
> > What happened?
> 
> Something went wrong... 
> 
> The end of base-Ex.Rout may contain a clue as to *what* went wrong.
> 
> If it was a crash of R, then the last few lines of output may be
> missing due to buffering and then it is more useful to run
> 
> ../../bin/R --vanilla < base-Ex.R

There were several warnings in base-Ex.Rout, which I will add at the end
of the mail, but they do not seem to be too serios to me.

/usr/applic/packages/bin/R --vanilla < base-Ex.Rout

ended after a while in

> ## Keywords: 'regression'.
> 
> 
> rm(list = ls(all = TRUE)); .Random.seed <- c(0,rep(7654,3))
> ###--- >>> `load' <<<----- Reload Saved Datasets
> 
>       ## alias         help(load)
> 
> ##___ Examples ___:
> 
> # save all data
> save(list = ls(), file= "all.Rdata")
> 
> # restore the saved values to the current environment
> load("all.Rdata")
Error in load("all.Rdata") : restore file corrupted -- no data loaded
Execution halted

Can anybody help me further?
Comments on the "library(mclust)-error" reported below are appreciated as well.
Perhaps this is independent of the "check"-problem?

Christian

PS: Additional information:

> version
         _                     
platform i586-pc-linux-gnulibc1
arch     i586                  
os       linux-gnulibc1        
system   i586, linux-gnulibc1  
status                         
major    1                     
minor    1.0                   
year     2000                  
month    June                  
day      15                    
language R                     

(SuSe Linux 5.2)

* In my work with this R-1.1.0 up to now everything worked fine except

> library(mclust)
/usr/applic/packages/R/lib/R/bin/R.bin: can't resolve symbol '__moddi3'
/usr/applic/packages/R/lib/R/bin/R.bin: can't resolve symbol '__divdi3'
Error in dyn.load(x, as.logical(local), as.logical(now)) : 
        unable to load shared library "/usr/applic/packages/R/lib/R/library/mclust/libs/mclust.so":
  Unable to resolve symbol
Error in library(mclust) : .First.lib failed

(This happened with R-0.64.0 on the same system as well, and also under 
Solaris. I assume that f2c has to be installed to prevent this, because
mclust works on the only machine in our institute where f2c is installed.)

* Warning messages in base-Ex-Rout:

>  summary(bY <- besselY(2,nu = nu <- seq(3,300,len=51)))
       Min.     1st Qu.      Median        Mean     3rd Qu.        Max. 
       -Inf        -Inf -2.248e+263        -Inf -3.777e+116  -1.128e+00 
There were 22 warnings (use warnings() to see them)

> dpois(c(0, 1, 0.17, 0.77), 1)
[1] 0.3678794 0.3678794 0.0000000 0.0000000
Warning messages: 
1: non-integer x = 0.170000 
2: non-integer x = 0.770000 

> curve(gamma(x),-3,4, n=1001, ylim=c(-10,100),
+       col="red", lwd=2, main="gamma(x)")
Warning message: 
NaNs produced in: gamma(x) 

> persp(tt,ncp,ptn, zlim=0:1, r=2, phi=20, theta=200, main=t.tit,
+       xlab = "t", ylab = "noncentrality parameter", zlab = "Pr(T <= t)")
Warning message: 
parameter "main" couldn't be set in high-level plot() function 

> hist(U <- rwilcox(N, m=4,n=6), breaks=0:25 - 1/2, border="red", col="pink",
+      sub = paste("N =",N))
Warning messages: 
1: parameter "sub" couldn't be set in high-level plot() function 
2: parameter "sub" couldn't be set in high-level plot() function 

> drop1(glm.D93, test="F")
Single term deletions

Model:
counts ~ outcome + treatment
          Df Deviance    AIC F value  Pr(F)
<none>          5.129 56.761               
outcome    2   10.581 58.214   2.126 0.2350
treatment  2    5.129 52.761   0.000 1.0000
Warning message: 
F test assumes quasipoisson family in: drop1.glm(glm.D93, test = "F") 

> has.VR <- require(MASS, quietly = TRUE)
Warning message: 
There is no package called `MASS' in: library(package, char = TRUE, logical = TRUE, warn.conflicts = warn.conflicts,  
(Comment: MASS has been installed in the meantime)

***

Christian Hennig
Fachbereich Mathematik - SPST
 (Schwerpunkt Mathematische Statistik und Stochastische Prozesse,
 Zentrum fuer Modellierung und Simulation)
Bundesstrasse 55
D-20146 Hamburg
Tel: x40/42838 4907, privat x40/631 62 79
hennig at math.uni-hamburg.de
http://www.math.uni-hamburg.de/home/hennig/



-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._