Prove that $\liminf [X_n = 2]) \subseteq ([\lim S_n] = \infty)$

39 Views Asked by At

Original problem:

Let $X_1, X_2, ...$ be random variables in $(\Omega, \mathscr{F}, \mathbb{P})$

where $X_1 = 0$ and for $n > 1$

$$P(X_n=-n^2)=\frac{1}{n^2}$$

$$P(X_n=-n^3)=\frac{1}{n^3}$$

$$P(X_n=2)=1-\dfrac{1}{n^2}-\dfrac{1}{n^3}$$

Let $S_n=\sum_{i=1}^{n} X_i$. Prove that $P([\lim S_n] = \infty) = 1$

1

There are 1 best solutions below

0
On BEST ANSWER

I used BCL1 to deduce that

$$P(\liminf [X_n = 2]) = 1$$

Let $\omega \in \Omega$. If $\omega \in \liminf [X_n = 2]$, then $\exists m \ge 1$ s.t.

$$2 = X_m(\omega) = X_{m+1}(\omega) = ...$$

$$\to S_{m+k}(\omega) = S_m(\omega) + 2k \ \forall k \ge 0$$

$$\to \lim _{k \to \infty} S_{m+k}(\omega) = \infty$$

$$\to \lim _{k \to \infty} S_{k}(\omega) = \infty$$

$$\to \lim _{n \to \infty} S_{n}(\omega) = \infty$$

The result follows by monotonicity of probability. QED