Gambler's ruin stopping time: can't understand the connection between the general and a specific solution

72 Views Asked by At

In Shiryaev's probability book, in page 87, it is given that

Taking account of $(8),$ we find that for $A \leq x \leq B$ $$ \beta(x)=\frac{(q / p)^{x}-(q / p)^{A}}{(q / p)^{B}-(q / p)^{A}}$$

[...]

If $p=q=\frac{1}{2},$ the only solutions $\beta(x)$ and $\alpha(x)$ of (7),(8) and (11),(12) are respectively $$ \beta(x)=\frac{x-A}{B-A} $$

However, if we put $p=q=0.5$ into the first equation, we simply get $(1-1)/(1-1)$, which is undefined, so I can't understand how does the author get the second equation from the first one.