What is the technical argument for this easy transformation?

26 Views Asked by At

In a book I found the following equation $$P(\forall s \in (0,h]~:~X_s \in D)=\inf_nP(\forall s\in (\frac{1}{n},h] ~:~X_s\in D).$$ where $X_s$ is some stochastic process starting in a set $D$. While it looked obvious to me at first, I have now asked myself what exactly is the technical argument behind this transformation. Is it some kind of continuity from above/below? Why did the author write $\inf$ instead of $\lim$?

1

There are 1 best solutions below

3
On BEST ANSWER

Let $E_t$ be the subset of the universe corresponding to $\forall s \in (t,h], X_s \in D$.

Then $E_0$ is the decreasing intersection of the $E_{1/n}$. The equation is simply monotone continuity for probabilities.