I have found some mixed messages in terms of the effective sample size to shoot for (ranging from 2,000 to 10,000) here.
I'm wondering if anyone could give me more information on choosing a guideline for effective sample size?
In particular, I am using a No-U-Turn Sampler.