Long-time behavior of multi-particle Markovian models (Q4801545)

Consider the sequence \(L(N)\), \(N= 1,2,\dots,\) where \(L(N)\) is a finite Markov chain on the state space \(X(N)\). Assume that \(|X(N)|\to\infty\) as \(N\to\infty\). Let \(P_N\) be a transition matrix of \(L(N)\), and let \(\mu\) denote an initial distribution of \(L(N)\). Assume that each \(L(N)\) is irreducible and aperiodic and denote by \(\pi^N\) the stationary distribution of \(L(N)\). A function \(T(N)\) is said to be a convergence time to equilibrium if for any function \(\psi(N)\uparrow\infty\) \[ \sup_\mu \|\mu P^{T(N)\psi(N)}_N- \pi^N\|\to 0\quad\text{as }N\to\infty. \] The paper gives the convergence time to equilibrium for wide classes of large multi-particle Markovian systems. It is shown that if a ``one-particle'' state space is large, then the long-time behavior of the multi-particle Markovian system strongly depends on the type of stochastic evolution of a single particle. For the entire collection see [Zbl 0995.00013].