lmBF (BayesFactor) - non-normal dependent variable
Hi
I'd like to compare two models, each including both categorical and continuous variables, using the BayesFactor package. From the documentation it seems lmBF is the function for this. But can lmBF still be used if the dependent variable isn't normal? and specifically if it can only take the values 1/0/-1? I couldn't seem to find any information regarding restrictions on the dependent variable.
I found this discussion:
https://forum.cogsci.nl/discussion/2360/can-i-use-lmbf-for-logistic-regression
which seems to indicate I can't use lmBF, but maybe this has changed? If lmBF isn't possible - are there any other suggestions?
Thanks
Comments
I don't think it can, at the moment --link functions need to be added, and the default prior distributions will need to change-- but I'll ask how Richard feels about this.
E.J.