Stability and accuracy of replication Bayes factor

wnk4242

Hello forum and Bayesian experts,

I was looking for published or unpublished studies that examine the stability and accuracy of replication BF. It seems that Verhagen and Wagenmakers have studied this in their 2014 study where they proposed replication BF. But I just wonder if there are any other simulation studies on this topic. Can anyone recommend some articles? Thank you!

Best,

EJ

Dear wnk4242,

I am not sure what you mean with "stability and accuracy", but note that a more recent paper on the topic is here: https://link.springer.com/article/10.3758/s13428-018-1092-x

Another one that may be relevant is here:

https://arxiv.org/abs/2207.14720

Cheers,

E.J.

wnk4242

Hi EJ,

Thanks for your response and info. Sorry if I didn't explain the two terms more clearly.

By "stability and accuracy", please see section 15.5.1 in this online post: https://vasishth.github.io/bayescogsci/book/ch-bf.html

Also, In Schad et al. (2022), they also mentioned stability and accuracy in Issue 1 and 2. Please see attached file.

Schad et al. - 2022 - Workflow techniques for the robust use of bayes fa.pdf

EJ

With respect to the instabilty: this is typical of *all* MCMC techniques -- so also for estimating the posterior mean of a parameter, say.

With respect to the bias: I have not encountered this as a problem (again, this concern seems to hold generally). But I should really look into that simulation method, looks pretty cool.

E.J.

wnk4242

Gotcha. Thanks again.