One of our experiments are getting the following a...
# ask-questions
One of our experiments are getting the following analysis error:
RegressionAdjustedStatistic cannot be used with the Bayesian statistics engine
I assume it's based on a custom metric the user added to the experiment, but the error doesn't specify which one it is (either does the stacktrace from the pod logs), before I go and try and binary search through the 40 or so metrics they've added, wondering if anyone knows off the top of their head what configuration on an experiment/metric is incompatible and would cause the analysis engine to throw this?
Hey, sorry for this. This is a bug on our end and I was able to reproduce it by clicking around a bit. I'll try and get a fix out soon. For more detail: basically if you have a default from the org or metric that RA should be on and try to run a bayesian analysis you'll hit this error, so I think a recent refactor of how we pass settings to our stats engine missed an override.
awesome, really appreciate the quick fix!