[R-sig-ME] lme4: Error from allFit when applied to lmer model fitted with changed control parameters

[Ben Bolker]

  I've posted this as an issue: https://github.com/lme4/lme4/issues/520

  There probably ought to be an `optCtrl` option to `allFit` that would
all optimizer-specific controls ...

On 2019-05-10 7:07 a.m., Gerrit Eichner wrote:
> Dear lme4-experts,
> 
> when fitting an lmer model with changed control parameters as in the
> following toy example
> 
> fm <- lmer(Reaction ~ Days + (Days | Subject), sleepstudy,
>            control = lmerControl(optCtrl = list(xtol_abs = 1e-8,
>                                                 ftol_abs = 1e-8)))
> 
> and then trying to refit the fitted model with all available optimizers
> using allFit an error is thrown:
> 
> allFit(fm)
> 
> bobyqa : Error in do.call(if (isGLMM(m)) glmerControl else lmerControl,
> ctrl) : second argument must be a list
> 
> 
> The problem originates apparently from the function ffun within allFit
> which doesn't seem to properly pass on the already provided control
> list.
> 
> Does anyone have a work-around for this? I would like to compare all
> optimizers with the changed optCtrl setting.
> 
> Thanks for any hints.
> 
>  Best regards  --  Gerrit
> 
>> sessionInfo()
> R version 3.5.3 (2019-03-11)
> Platform: x86_64-w64-mingw32/x64 (64-bit)
> Running under: Windows 10 x64 (build 17763)
> 
> Matrix products: default
> 
> locale:
> [1] LC_COLLATE=German_Germany.1252  LC_CTYPE=German_Germany.1252
> [3] LC_MONETARY=German_Germany.1252 LC_NUMERIC=C
> [5] LC_TIME=German_Germany.1252
> 
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
> 
> other attached packages:
> [1] dfoptim_2018.2-1 lme4_1.1-21      Matrix_1.2-17
> 
> loaded via a namespace (and not attached):
>  [1] minqa_1.2.4     MASS_7.3-51.3   compiler_3.5.3  Rcpp_1.0.1
>  [5] splines_3.5.3   nlme_3.1-137    grid_3.5.3      nloptr_1.2.1
>  [9] boot_1.3-20     lattice_0.20-38
> 
> 
> ---------------------------------------------------------------------
> Dr. Gerrit Eichner                   Mathematical Institute, Room 212
> gerrit.eichner using math.uni-giessen.de   Justus-Liebig-University Giessen
> Tel: +49-(0)641-99-32104          Arndtstr. 2, 35392 Giessen, Germany
> http://www.uni-giessen.de/eichner
> 
> _______________________________________________
> R-sig-mixed-models using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models