Bayes Factor as non-logistic output from proportionBF

74 views Asked by At

I am estimating a BayesFactor of proportions using the comand proportionBF() from R. I want to replicate the example from wikipedia https://en.wikipedia.org/wiki/Bayes_factor. The output from the proportionBF comand is in log, so I transform the output of log trying to coincide results with the example of wikipedia, by I am making a mistake. My code is:

library(BayesFactor)
bf = proportionBF(115, 200, p = 1/2)
exp(bf@bayesFactor$bf)

Can anybody help me to understand which is my miskate.

1

There are 1 answers

0
Victor19 On BEST ANSWER

According to the R document (You can find it by typing ?proportionBF).

the default alternative is that λ~logistic(λ_0,r)

where lambda_0=logit(p_0) and lambda=logit(p). r serves as a prior scale parameter.

But from Wiki, the prior distribution for q is coming from uniform distribution.