Taking the average of replication Bayes factors
Hello,
I am doing a simulation study to evaluate some statistical properties of replication bayes factor. Suppose I simulate an original study and then simulate a number of exact replication studies (e.g., n=5) based on the original study's parameters and obtain replication bayes factors for each replication study. Is it a good idea to take the average of the replication bayes factors to obtain a mean replication bayes factor?
I understand that there could be outliers and taking the average when outliers exist may not be a good idea. If so, does the median work? Or should I first take the log of the replication bayes factors and then take the average of the logBFs? Or are there other ways to obtain an average replication BF?
Thank you!
Comments
Taking the average of BFs in generally inadvisable. For instance, the average of 3 and 1/3 is not 1.
So taking the mean of the log BFs is indeed a better idea.
E.J.
Thank you, EJ!