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
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.
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.