Skip to content

Problems running lmer

4 messages · Andrea Goijman, Philippi, Tom, Hugo.Mildenberger at web.de

#
Andrea, Tom,

I was able to reproduce the problem here. Apart from an unfortunate model specification, the technical reason is a stack overflow which sometimes does not even result in an error message (the program gets killed instead). From the kernel logs I can see that R tried to allocate 13 mega bytes of stack space. Using gdb on the core dump shows that this must have happened in function mer_optimize() at lmer.c:1719. 
 

   > # varying intercept by route, species fixed effect
   > m1 <-  lmer(cbind(punto6,5) ~ sp + (1|route) ,family=binomial,data = d)

   > # varying intercept by site within route, species fixed effect
   > m2 <-  lmer(cbind(punto6,5) ~ sp + (site|route) ,family=binomial,data = d)

   Error: segfault from C stack overflow

   > sessionInfo()
   R version 3.0.0 (2013-04-03)
   Platform: x86_64-pc-linux-gnu (64-bit)

   [...]

   other attached packages:
   [1] lme4_0.999999-2 Matrix_1.0-12   lattice_0.20-15

   loaded via a namespace (and not attached):
   [1] grid_3.0.0   nlme_3.1-109 stats4_3.0.0


Best





On Fri, 17 May 2013 16:34:07 -0400
Andrea Goijman <agoijman at cnia.inta.gov.ar> wrote: