[R-sig-ME] parallel option of the bootMer function: warnings for parallel=no, but not for parallel=snow

Spronken, M. (Maitta) m.spronken at psych.ru.nl
Mon Oct 27 14:20:18 CET 2014

Dear all,

Last week I asked two questions about the parallel option of the bootMer function. As I explained in my first post, when bootstrapping my model, I got warnings when using parallel="no", but not when using parallel="snow". Ben Bolker suggested that some of the internal structures of lme4 might be getting reset properly when using refit() with parallel="snow", but not with parallel="no".

As I explained in my second post, whether I get warnings when using parallel="no" also seems to depend on the set.seed() value chosen and whether I run my code on a Mac/Windows pc. I asked whether someone knows what this indicates, but maybe I should ask my question more specifically: 

Do you have an idea whether different set.seed() values and using Mac vs. Windows leads to differences in the occurrence of warnings because parallel="no" might not reset some internal structures properly, or could these differences also indicate that there is something 'wrong' with my data/model?

Thanks in advance for any suggestions!


More information about the R-sig-mixed-models mailing list