Let $X_t=2+t+W_t$ for $t\ge 0$, where $(W_t)_{t\ge 0}$ is a standard Brownian motion. For every $n\ge 1$, set $X^n_t:=X_t-{\bf 1}_{t\ge n}$. Denote respectively

$$\tau:=\inf\{t\ge 0:~ X_t\le 0\}\quad \mbox{and} \quad \tau^n:=\inf\{t\ge 0:~ X^n_t\le 0\}.$$

Could we prove or disprove $\lim_{n\to\infty}\mathbb P[\tau^n=\infty]=\mathbb P[\tau=\infty]$?

Further Question : I wish to prove the similar convergence result. This question can be found at Convergence of the probabilities that drifted Brownian motion with jump never hits zero (continuation)

  • $\begingroup$ The equality is an immediate consequence of the fact that $\tau^n \uparrow \tau$ a.s. Note that $X_t \to \infty$ a.s. $\endgroup$ – Dieter Kadelka Jun 1 at 22:00
  • $\begingroup$ @DieterKadelka I got the point. Now I write down the claim that I wish to prove (see the post above). Do you think the result is still true? $\endgroup$ – Neymar Jun 1 at 22:38
  • $\begingroup$ Hello @Neymar , you should ask this in a new question. $\endgroup$ – Dieter Kadelka Jun 2 at 8:58
  • $\begingroup$ @DieterKadelka Thanks. I post the new question at mathoverflow.net/questions/394332/… $\endgroup$ – Neymar Jun 2 at 9:11

Just use the Markov property at time $n$.

Write $f(x)$ for the probability that $x+t+W_t$ never hits $(-\infty, 0]$. Then $$ \mathbb P[\tau = \infty] = \mathbb E[\mathbb 1_{\{\tau > n\}} f(X_n)] $$ and $$ \mathbb P[\tau^n = \infty] = \mathbb E[\mathbb 1_{\{\tau > n\}} f(X_n - 1)] , $$ so the difference of the two is $$ \mathbb E[\mathbb 1_{\{\tau > n\}} (f(X_n) - f(X_n - 1))] . $$ Now $X_n \to \infty$ and $\mathbb 1_{\{\tau > n\}} \to \mathbb 1_{\{\tau = \infty\}}$ with probability one as $n \to \infty$, and $f(x) - f(x - 1) \to 0$ as $x \to \infty$. Thus, the integrand converges to zero with the probability one, and the dominated convergence theorem implies that indeed $$ \mathbb P[\tau = \infty] - \mathbb P[\tau^n = \infty] \to 0 . $$

  • $\begingroup$ Very nice reasoning! My question aims to find a counterexample for some claim that I am unable to prove. Now I have written down my claim. Do you think it is still true? $\endgroup$ – Neymar Jun 1 at 22:34
  • $\begingroup$ This claim has been rewritten in a new post, see mathoverflow.net/questions/394332/… $\endgroup$ – Neymar Jun 2 at 9:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.