[R-sig-ME] Extracting variances from a mer object

Gang Chen gangchen6 at gmail.com
Wed Jul 7 23:18:54 CEST 2010


Thanks a lot for the quick help! I really appreciate it...

Gang

On Wed, Jul 7, 2010 at 5:16 PM, Douglas Bates <bates at stat.wisc.edu> wrote:
> On Wed, Jul 7, 2010 at 4:13 PM, Douglas Bates <bates at stat.wisc.edu> wrote:
>> On Wed, Jul 7, 2010 at 4:06 PM, Gang Chen <gangchen6 at gmail.com> wrote:
>>> This is a simple question. When fitting a model with lmer, how can I
>>> extract all the variances from the output object? For example, with
>>> the following analysis, I just want to extract the column of Variance
>>> under the Random effects table - those 3 variance numbers: 0.63756 for
>>> Subj, 23.03698 for Scan, and 199.50496 for Residual. VarCorr(fm) only
>>> gives me the first two. I couldn't find any slots for this purpose
>>> from str(fm) either.
>>
>> ?VarCorr
>
> Sorry, I answered too quickly.  You had checked the result of VarCorr.
>  The residual variance is square of the "sc" attribute.
>
>> fm1 <- lmer(Yield ~ 1 + (1|Batch), Dyestuff)
>> str(VarCorr(fm1))
> List of 1
>  $ Batch: num [1, 1] 1764
>  ..- attr(*, "dimnames")=List of 2
>  .. ..$ : chr "(Intercept)"
>  .. ..$ : chr "(Intercept)"
>  ..- attr(*, "stddev")= Named num 42
>  .. ..- attr(*, "names")= chr "(Intercept)"
>  ..- attr(*, "correlation")= num [1, 1] 1
>  .. ..- attr(*, "dimnames")=List of 2
>  .. .. ..$ : chr "(Intercept)"
>  .. .. ..$ : chr "(Intercept)"
>  - attr(*, "sc")= num 49.5
>> fm1
> Linear mixed model fit by REML
> Formula: Yield ~ 1 + (1 | Batch)
>   Data: Dyestuff
>   AIC   BIC logLik deviance REMLdev
>  325.7 329.9 -159.8    327.4   319.7
> Random effects:
>  Groups   Name        Variance Std.Dev.
>  Batch    (Intercept) 1764.0   42.001
>  Residual             2451.3   49.510
> Number of obs: 30, groups: Batch, 6
>
> Fixed effects:
>            Estimate Std. Error t value
> (Intercept)  1527.50      19.38   78.81
>
>
>>>> (fm<-lmer(Rep~(1|Subj)+(1|Scan), data=Model))
>>> Linear mixed model fit by REML
>>> Formula: Rep ~ (1 | Subj) + (1 | Scan)
>>>   Data: Model
>>>   AIC   BIC logLik deviance REMLdev
>>>  166.4 170.3 -79.18    163.2   158.4
>>> Random effects:
>>>  Groups   Name        Variance  Std.Dev.
>>>  Subj     (Intercept)   0.63756  0.79847
>>>  Scan     (Intercept)  23.03698  4.79969
>>>  Residual             199.50496 14.12462
>>> Number of obs: 20, groups: Subj, 10; Scan, 2
>>>
>>> Fixed effects:
>>>            Estimate Std. Error t value
>>> (Intercept)   27.603      4.642   5.946
>>>
>>> Thanks,
>>> Gang
>>>
>>> _______________________________________________
>>> R-sig-mixed-models at r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>>>
>>
>



-- 
Existence is elsewhere.
— André Breton, “The Surrealist Manifesto”




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