mschauer / ZigZagBoomerang.jl

Sleek implementations of the ZigZag, Boomerang and other assorted piecewise deterministic Markov processes for Markov Chain Monte Carlo including Sticky PDMPs for variable selection
MIT License
101 stars 7 forks source link

Should the iterators give the initial state as first return value? #85

Open mschauer opened 3 years ago

ParadaCarleton commented 3 years ago

Assuming the iterators don't discard burn-in, I assume so; it might be useful for something and it can't really hurt. If the iterators discard burn-in, the initial state should probably be discarded, since most people will assume that samples at the start should represent the posterior distribution.