Consider the transition matrix
$ P = \begin{bmatrix} 1-p&p\\ q&1-q \end{bmatrix} $
for general $2$-state Markov Chain $(0 \le p, q\le 1)$.
- Find the limiting distribution (if it exists) if $p + q \ne 1$.
Using mathematical induction, it is solved using the following proof:
$ p^n = \frac{1}{p+q} \begin{bmatrix}q&p\\q&p\end{bmatrix} + \frac{(1-p-q)^n}{p+q} \begin{bmatrix}p&-p\\-q&q\end{bmatrix} $
This is totally cumbersome.
Can this be solved in any other way like using $\pi (P-I) = 0$ and so on?
I think induction works well as long as you don't make a dumb mistake on the matrix math (I did the first time I tried it!).
I apologize for the formatting!! Still getting used to latex.
For the first term, we can show that it stays the same after being multiplied by the transition matrix:
$\begin{bmatrix} q&p \\q&p \end{bmatrix} \begin{bmatrix} 1-p & p \\ q & 1-q \end{bmatrix} = \begin{bmatrix} q(1-p) + pq & pq+p(1-q) \\ q(1-p) + pq & pq+p(1-q) \end{bmatrix} $
$\begin{bmatrix} q&p \\q&p \end{bmatrix} \begin{bmatrix} 1-p & p \\ q & 1-q \end{bmatrix} = \begin{bmatrix} q-qp + pq & pq+p-pq) \\ q-qp + pq & pq+p-pq) \end{bmatrix} $
$\begin{bmatrix} q&p \\q&p \end{bmatrix} \begin{bmatrix} 1-p & p \\ q & 1-q \end{bmatrix} = \begin{bmatrix} q & p \\ q & p \end{bmatrix}$
For the second term, it's a little tougher:
$\begin{bmatrix} p & -p \\ -q & q \end{bmatrix} \begin{bmatrix} 1-p & p \\ q & 1-q \end{bmatrix} = \begin{bmatrix} p(1-p) - pq & p^2-p(1-q) \\ -q(1-p) -q^2 & -qp+q(1-q) \end{bmatrix}$
$\begin{bmatrix} p & -p \\ -q & q \end{bmatrix} \begin{bmatrix} 1-p & p \\ q & 1-q \end{bmatrix} = \begin{bmatrix} p(1-p-q) & -p(1-q-p) \\ -q(1-p-q) & q(1-q-p) \end{bmatrix}$
$\begin{bmatrix} p & -p \\ -q & q \end{bmatrix} \begin{bmatrix} 1-p & p \\ q & 1-q \end{bmatrix} = (1-p-q) \begin{bmatrix} p & -p \\ -q & p \end{bmatrix}$
this creates the $(1-p-q)^{n+1}$ term
