Extremely high BFIncl in RM ANOVA
Hi all,
I conducted a Bayesian RM ANOVA, including the factor TIME (4 levels) and GROUP (3 levels). However, I observed some extremely high BFIncl, e.g., of 2.541e+10 for the main effect of TIME.
Is that possible, like that is the real value, or could there be something wrong?
Have any other also observed such extreme BFIncl?
Many Thanks!
Nils
Comments
Definitely possible - Bayes factors are ratios, and they can be arbitrarily larger than 1, or arbitrarily smaller than 1.
E.g., :
<code>
library(BayesFactor)
BF <- generalTestBF(len ~ supp * dose, ToothGrowth)
bayestestR::bayesfactor_inclusion(BF)
#> # Inclusion Bayes Factors (Model Averaged)
#>
#> Pr(prior) Pr(posterior) Inclusion BF
#> supp 0.6 0.981 35.206
#> dose 0.6 1.000 5.774e+12
#> dose:supp 0.2 0.559 5.067
#>
#> * Compared among: all models
#> * Priors odds: uniform-equal
</code>
Okay, thanks for your quick answer and the code!
Best
Nils