In extreme cases, where an interval hypothesis is highly consistent with the data, the Bayes factor returned is NaN:
> t = c(-.15,2.39,2.42,2.43)
> N = c(100,150,97,99)
> meta.ttest.tstat(t,n1=N,nullInterval = c(0,Inf),rscale=1)
$bf
null alt
3.65618 NaN
$properror
null alt
0.0001029407 0.1725509594
The following warning is given:
Warning message:
In log(complArea) : NaNs produced
This is caused by the extremely small amount of posterior area for the hypothesis in question.
In extreme cases, where an interval hypothesis is highly consistent with the data, the Bayes factor returned is
NaN:The following warning is given:
This is caused by the extremely small amount of posterior area for the hypothesis in question.