[R] R bug on survival library Windows 7

[Ben Bolker]

CASTELLI christel <christel.CASTELLI <at> chu-nimes.fr> writes:

> 
> Dear All,
> 
> I have updated my OS - Windows 7, and I found a 
> situation where R stops working
> with reported error "R for Windows GUI front-end has 
> stopped working".  My script worked before (on windows XP).
> 
> The error occured when I perform this:
> 
> coxph(Surv(time,event)~F11_rs2036914+frailty(n), data=base)
> 
> from survival library.
> 
> If I delete the specification frailty(n) this works! But I have clustered
data and need this specification.
> 
> Here is my system information:
> R version 3.0.2 (2013-09-25)
> Platform: x86_64-w64-mingw32/x64 (64-bit)
> locale:
> [1] LC_COLLATE=French_France.1252  LC_CTYPE=French_France.1252
> [3] LC_MONETARY=French_France.1252 LC_NUMERIC=C
> [5] LC_TIME=French_France.1252
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
> 
> Do someone meet this problem?
> 
> Anyone has some idea how to avoid this crash in Windows 7 system?
> 
> Note that I can't go back on windows XP because 
> my institution do the choice of windows 7.
> 
> Sincerely,
> 
> Christel Castelli, PhD
> Ingénieur biostatisticien
> Responsable de la cellule médico-économie
> Département BESPIM
> CHU de Nîmes
> 
 
  You may get lucky and discover that someone else has seen a very
similar symptom, but it is very most likely that you will have to
produce a reproducible example <http://tinyurl.com/reproducible-000>.
That means that (1) you provide data, or simulated data, that
(2) reliably produces this crash in a clean session on your machine
(for some kinds of bugs the best you can do may be "if I run this
1000 times in succession it almost always crashes") (3) can produce
this crash on someone else's machine (same version of R, same OS,
perhaps the same locale).

  Googling for this problem suggests it may be a memory exhaustion
issue: http://r.789695.n4.nabble.com/
  r-for-windows-gui-front-end-has-stopped-working-td4666659.html

(broken URL)

  How much memory do you have, how big is your data set?