Intuition behind invariant measure for a stochastic matrix Norris Theorem 1.7.6

83 Views Asked by At

I'm trying to figure out the meaning of the following theorem from J. Norris - Markov Chains.

Let $$\gamma_i^k=E_k\sum_{n=0}^{T_k-1}1_{\{X_n=i\}}$$ Let $P$ be irreducible and let $\lambda$ be an invariant measure for $P$ with $\lambda_k=1$.

Then $\lambda \ge \gamma^k$. If in addition $P$ is recurrent, then $\lambda=\gamma^k$.

The way I see it, this means that if there's an invariant measure for which $\lambda_k=\gamma_k^k=1$, then this invariant measure $\lambda$ is definitely greater or equal to the $\gamma^k$ invariant measure.

I'm not looking for a proof, but rather an understanding of why the following theorem holds true. What's the intuitive way to think about it?

Thanks in advance!