[R-sig-ME] lmer formula syntax?

[Douglas Bates]

On Wed, Feb 2, 2011 at 3:43 PM, Colin Wahl <biowahl at gmail.com> wrote:
> I too have found the Bate's book draft helpful, but I have not been able to
> get the lme4a package installed. The best description I have found to do so
> is:
> http://kmyu.wordpress.com/2010/12/04/installing-lme4a-in-r-mac-osx-10-6-4/

> I'm understanding that the lme4a package is dependent on a number of
> packages: RcppArmadillo, minqa and MatrixModels (available on CRAN
> respositories) and Rcpp version 0.8.8.1 or later (which is not; version
> 0.8.6 is).

(Actually it doesn't depend on RcppArmadillo).

You must be using Mac OS X.  If you check at
cran.r-project.org/web/packages/Rcpp you will see that the Windows
binary and the source code packages are at version 0.9.0 but the Mac
OS X version is only at 0.8.6, which is rather old.  I'm not sure what
the problem is as Romain develops on a Mac OS X system and he can't
reproduce the error.

However, you don't really need lme4a for most of the examples in the
book.  The lme4 package can't handle the profile-related material, but
the rest of the examples should be okay.  The output looks slightly
different but the parameter estimates should be essentially the same.





> Apparently it is available on something called an SVN
> respository. Im getting into language I am unfamiliar with. If you can
> provide some advice to aid my ignorance in installing this package I could
> likely get a lot more out of the book draft.
>
> Thank for you time and assistance.
>
> On Tue, Feb 1, 2011 at 6:34 PM, George Wang <pseudotelphusa at gmail.com>wrote:
>
>> I am also quite new to lme4 myself, so I'll just add one thing to Toby's
>> earlier answer.
>>
>> On Tue, Feb 1, 2011 at 5:14 PM, Colin Wahl <biowahl at gmail.com> wrote:
>>
>>> Dear r-sig-mixed-models List,
>>>
>>> [snip]
>>
>>
>>
>>> Secondarily: I have read a few times that "wshed/stream" is
>>> interchangeable
>>> with  "wshed:stream" which is a meaningless interaction. Also I've seen
>>> that
>>> random effects are specified as (a|b) where a is a covariate and b is a
>>> grouping factor. Does having 1 as a covariate simply specifying an
>>> intercept
>>> of 1? What is the purpose of placing a factor in the place of 1 as a
>>> covariate? OR is there a nice complete summary tutorial that I've missed.
>>>
>> A specification of (1|b) means that the model is fit with a random effect
>> for each level of b (simple scalar random effect). The "a" in (a|b) is a
>> correlated random effect along with "b", with both an intercept and slope.
>> You can also make them uncorrelated, with slope only for a.
>>
>> The draft of Prof. Doug Bate's book covers many useful basics on lmer, and
>> is where I learned most of mine.
>> http://lme4.r-forge.r-project.org/book/
>>
>>
>
>
>> HTH
>> George
>>
>>
>>>
>>> I'm looking forward to hearing any comments.
>>>
>>> Thank you,
>>> --
>>> Colin Wahl
>>> Department of Biology
>>> Western Washington University
>>> Bellingham, WA 98225
>>> ph: 360-391-9881
>>>
>>>        [[alternative HTML version deleted]]
>>>
>>>
>>> _______________________________________________
>>> R-sig-mixed-models at r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>>>
>>>
>>
>
>        [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>