Trace plot
Topic of Markov chains, Gibbs sampling, and Metropolis–Hastings, part of Bayesian statistics
Trace plot
| Type | Topic |
| Taught in | Markov chains, Gibbs sampling, and Metropolis–Hastings |
| Part | Bayesian statistics |
| Questions using it | 3 |
A plot of the sampled values of a parameter against iteration number. A chain that drifts, sticks, or disagrees with other chains shows a problem; a tidy plot is evidence of mixing, not a proof of convergence.
In Markov chains, Gibbs sampling, and Metropolis–Hastings
The trace plot and autocorrelation function are not substitutes for the stationarity argument. They address the separate practical question of whether this finite run appears to have explored the target well enough for the posterior summary at hand.
Used in
- Taught in Markov chains, Gibbs sampling, and Metropolis–Hastings
- Questions that use it (3) — Q19 (answer) · Question 3 (answer) · Question 16 (answer)
- All 31 questions on Markov chains, Gibbs sampling, and Metropolis–Hastings
See also
Markov chain · Stationarity and detailed balance · Detailed balance implies stationarity · Systematic-scan Gibbs sampler · Warm-up is not a convergence proof · Metropolis–Hastings update · Metropolis–Hastings invariance · Markov chains and equilibrium