[R-meta] negative AIC

[Lukasz Stasielowicz]

Dear Catia,

AIC values can be negative. In fact, values approaching negative 
infinity are better than values close to zero.

See for example this discussion for more details: 
https://stats.stackexchange.com/questions/84076/negative-values-for-aic-in-general-mixed-model/84077


Best,
Lukasz
-- 
Lukasz Stasielowicz
Osnabrück University
Institute for Psychology
Research methods, psychological assessment, and evaluation
Seminarstraße 20
49074 Osnabrück (Germany)

Am 05.11.2021 um 12:00 schrieb r-sig-meta-analysis-request using r-project.org:
> Send R-sig-meta-analysis mailing list submissions to
> 	r-sig-meta-analysis using r-project.org
> 
> To subscribe or unsubscribe via the World Wide Web, visit
> 	https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis
> or, via email, send a message with subject or body 'help' to
> 	r-sig-meta-analysis-request using r-project.org
> 
> You can reach the person managing the list at
> 	r-sig-meta-analysis-owner using r-project.org
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of R-sig-meta-analysis digest..."
> 
> 
> Today's Topics:
> 
>     1. negative AIC (=?UTF-8?Q?C=C3=A1tia_Ferreira_De_Oliveira?=)
>     2. rma.mv-When a higher level can't be modeled because of one
>        row (Farzad Keyhan)
>     3. Re:  rma.mv-When a higher level can't be modeled because of
>        one row (Farzad Keyhan)
> 
> ----------------------------------------------------------------------
> 
> Message: 1
> Date: Thu, 4 Nov 2021 18:04:27 +0000
> From: =?UTF-8?Q?C=C3=A1tia_Ferreira_De_Oliveira?= <cmfo500 using york.ac.uk>
> To: R meta <r-sig-meta-analysis using r-project.org>
> Subject: [R-meta] negative AIC
> Message-ID:
> 	<CACw+TfeuPABNv=Q9QzLgRBG17N3VAhgaHADkr4h1Ukyw=DCKPg using mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
> 
> Hello,
> 
> For a meta-analysis I am running I was planning on using the last model
> (res.ml2) but when comparing the models the AIC is negative. Is that a
> problem?
> 
> 
> res.ml0 <- rma.mv(yi, vi, random = ~ 1 | Study, data= Data)
> res.ml1 <- rma.mv(yi, vi, random = ~ 1 | Study/obs, data=Data)
> res.ml2 <- rma.mv(yi, vi, random = ~ 1 | Study/exp/obs, data=Data)
> 
> # Compare models
> anova(res.ml0, res.ml1)
> 
> 
>          df    AIC    BIC   AICc logLik    LRT   pval      QE
> Full     3 1.7400 5.6276 2.7835 2.1300               64.3868
> Reduced  2 2.8124 5.4041 3.3124 0.5938 3.0724 0.0796 64.3868
> anova(res.ml1, res.ml2)
> 
> anova(res.ml1, res.ml2)
> 
>          df     AIC     BIC    AICc logLik     LRT   pval      QE
> Full     4 -6.5959 -1.4125 -4.7777 7.2979                64.3868
> Reduced  3  1.7400  5.6276  2.7835 2.1300 10.3359 0.0013 64.3868
> 
> Best wishes,
> 
> Catia
>