Paper 1, Section II, 20H

Markov Chains | Part IB, 2014

Consider a homogeneous Markov chain (Xn:n0)\left(X_{n}: n \geqslant 0\right) with state space SS and transition matrixP=(pi,j:i,jS)\operatorname{matrix} P=\left(p_{i, j}: i, j \in S\right). For a state ii, define the terms aperiodic, positive recurrent and ergodic.

Let S={0,1,2,}S=\{0,1,2, \ldots\} and suppose that for i1i \geqslant 1 we have pi,i1=1p_{i, i-1}=1 and

p0,0=0,p0,j=pqj1,j=1,2,,p_{0,0}=0, p_{0, j}=p q^{j-1}, j=1,2, \ldots,

where p=1q(0,1)p=1-q \in(0,1). Show that this Markov chain is irreducible.

Let T0=inf{n1:Xn=0}T_{0}=\inf \left\{n \geqslant 1: X_{n}=0\right\} be the first passage time to 0 . Find P(T0=nX0=0)\mathbb{P}\left(T_{0}=n \mid X_{0}=0\right) and show that state 0 is ergodic.

Find the invariant distribution π\pi for this Markov chain. Write down:

(i) the mean recurrence time for state i,i1i, i \geqslant 1;

(ii) limnP(Xn0X0=0)\lim _{n \rightarrow \infty} \mathbb{P}\left(X_{n} \neq 0 \mid X_{0}=0\right).

[Results from the course may be quoted without proof, provided they are clearly stated.]

Typos? Please submit corrections to this page on GitHub.