The developer of the well-known
emcee package often gives this advice to help with chain convergence:
- Run a short (few hundred steps) chain
- Reinitialize all the walkers near the point with maximum log probability seen so far
- Return to step 1 a few times
- Then run your final chain starting where you ended up for your last run of step 1
My question is: is this any different than simply allowing the chain(s) to run for more steps and then discard a longer burn-in period? I.e., is this any different from a "normal" run with more burn-in steps?