Markov Chain

overview

Summary

A Markov chain tracks how likely a system is to move from one state to another. The transition data can be stored either as an adjacency list (per-state neighbors and weights) or as an adjacency/transition matrix (tabular probability grid). The key idea is that the next state depends only on the current state - this is the Markov property. Over repeated transitions, the chain can settle into stable long-term behavior known as the stationary distribution.