Metropolis-Hasting involves designing a Markov process by constructing Transition Probabilities, which has a unique stationary distribution if it fulfills two conditions:
- Existitence of Stationary Distribution:
Reason----Detailed Balance:
- Uniqueness of Stationary Distribution:
Reason – Ergodicity: every states satisfies
1 Aperiodic: the system does not return to the same state at fixed intervals----
2 Be positive recurrent - the expected number of steps for returning to the same state is finite.
The derivation of the algorithm starts with the condition of detailed balance.
, where A(x’, x) is the acceptance ratio, the probability to accept the proposed state x’.
The next step in the derivation is to choose an acceptance ratio that fulfils the condition above. One common choice is the Metropolis choice:
reference
https://en.wikipedia.org/wiki/Metropolis–Hastings_algorithm