[R-sig-ME] lmer

Douglas Bates bates at stat.wisc.edu
Fri Jun 18 16:46:46 CEST 2010


On Fri, Jun 18, 2010 at 8:42 AM, Ricardo Alvarado
<hipotesis_delta at hotmail.com> wrote:
> Thank you very much for your response.
>
> I really need to use this package and make it work, maybe you can help me to
> define the right model I need to input.
>
> I have 3 blocks with 5 treatments within each block.  We gathered
> information from response variables at 5 different times on the same units -
> so I have a longitudinal study.
>
> If I call:
>
> B: blocks        - a,b,c  (factor)
> T: treatments  - A,B,C,D,E  (factor)
> t: time            -  15, 30, 45, 60, 75 days (continuous)
>
> How do I write the model and how do I determine the degrees of freedom?

Please send such messages to the R-SIG-Mixed-Models at R-project.org
mailing list, which I am cc:ing on this reply.

>
> I wrote this:
>
>            lmer(y~t*T+(t|B))
>
> but I am not sure if it is taking into accout the random nature of the
> blocks where the treatements are included.
>
> I would really appreciate if you could give me some insight on this.
>
> Best regards,
>
> Ricardo
> ----------------------
>
> On Thu, Jun 17, 2010 at 3:56 AM, Ricardo Alvarado
> <hipotesis_delta at hotmail.com> wrote:
>> Hello:
>> I am using package lme4. When I run a model with lmer I don't get the
>> p-values.
>
> That's intentional.  See R FAQ (Frequently Asked Questions) 7.35
>
> 7.35 Why are p-values not displayed when using lmer()?
>
> Doug Bates has kindly provided an extensive response in a post to the
> r-help list, which can be reviewed at
>
> https://stat.ethz.ch/pipermail/r-help/2006-May/094765.html.
>
>> Is there any way to get them directly without have to compute from
>> the F values?
>
> Before you can evaluate a p-value from an F distribution you will need
> to decide on the degrees of freedom in the denominator and that is the
> source of controversy.
>
> ________________________________
> Hotmail es más seguro. Con el Filtro Anti Spam tu cuenta está mucho más
> protegida. Ver más




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