[R-sig-ME] lme4 1.1-21 going to CRAN soon

Ben Bolker bbo|ker @end|ng |rom gm@||@com
Wed Feb 27 00:01:19 CET 2019


  The CRAN maintainers have asked for another lme4 update to fix some
testing issues. We've also made the following few changes

  USER-VISIBLE CHANGES:

   • ‘bootMer’ now traps and stores messages, warnings, and errors
   • ‘bootMer’ returns an object of class ‘c("bootMer","boot")’; new
‘print’ and ‘confint’ methods for class ‘bootMer’
   • small changes to wording of singular-fit messages

---

We don't anticipate any downstream problems (we have run
reverse-dependency checks -- but we have missed problems before!), but
would be happy to hear about any problems, or relatively straightforward
bug fixes/feature requests that we should try to squeeze into the next
CRAN release (CRAN rules policies discourage us from submitting new
releases more frequently than every 1-2 months).

 [In case you're thinking of asking, "implement flexible correlation and
heteroscedasticity structures at the level of residuals and groups" is
not relatively straightforward :-( ]

  cheers
    Ben Bolker



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