Message-ID: <AANLkTimRScV6_mKTx_g44AvFaUAty_ym9_-k1JwuckNb@mail.gmail.com>
Date: 2010-12-29T17:35:45Z
From: Gabor Grothendieck
Subject: as.object: function doesn't exist but I wish it did
In-Reply-To: <AANLkTimT3F0q3Q6D1Pe6nVryU96QGtjmu0B1vAYKaCQS@mail.gmail.com>
On Wed, Dec 29, 2010 at 12:31 PM, Patrick McKann <pcmckann at gmail.com> wrote:
> I seem to come to this problem alot, and I can find my way out of it with a
> loop, but I wish, and wonder if there is a better way. ?Here's an example
> (lmer1-5 are a series of lmer objects):
>
> ?bs=data.frame(bic=BIC(lmer1,lmer2,lmer3,lmer4,lmer5)$BIC)
> ?rownames(bs)=c('lmer1','lmer2','lmer3','lmer4','lmer5')
> ?best=rownames(bs)[bs==min(bs)]
>
>> best
> [1] "lmer5"
>
> This tells me that lmer5 is the model with the lowest BIC. ?I want to start
> working with lmer5 as the best model, such as fixef(best) to get the fixed
> effect estimates from lmer5. ?I tried best=as.object('lmer5') but of course
> this doesn't work because that is not a real function.
>
> Does anybody see what I'm getting at? ?If so, do you know a way to do this
> without a loop or series of if statements?
>
If you are asking how to turn a string into a variable its a FAQ!
http://cran.r-project.org/doc/FAQ/R-FAQ.html#How-can-I-turn-a-string-into-a-variable_003f
--
Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com