I am stuck on Problem 4.12 of Karatzas and Shreve's book Stochastic Calculus and Brownian Motion:

Suppose that $\{ \mathbb{P}_n \}$ is a sequence of probability measures on $(C[0, \infty), \mathcal{B} (C[0, \infty)))$ which converges weakly to a probability measure $\mathbb{P}$. Suppose, in addition, that $\{f_n\}$ is a uniformly bounded sequence of real-valued, continuous functions on $C[0, \infty)$ converging to a continuous function $f$, the convergence being uniform on compact subsets of $C[0, \infty)$. Then $$ \lim_{n \to \infty} \int_{C[0,\infty)} f_n \, d \mathbb{P_n} = \int_{C[0,\infty)} f \, d \mathbb{P} .$$

The result is clear if the iterated limit is considered, i.e. $$ \lim_{m \to \infty} \lim_{n \to \infty} \int_{C[0,\infty)} f_n \, d \mathbb{P_m} = \int_{C[0,\infty)} f \, d \mathbb{P} .$$

But I am not sure how we could show this statement. Any ideas?

  • $\begingroup$ Hint 1: use the unfimor convergence of $f_n$ Hint 2: dominated convergence $\endgroup$ – noctusraid Feb 5 '16 at 17:41
  • $\begingroup$ @noctusraid But uniform convergence only holds for compact subsets. So how can we approximate $C[0, \infty)$ by its compact subsets??? $\endgroup$ – Richard Feb 5 '16 at 17:44
  • $\begingroup$ Remember that the convergent sequence $\{\Bbb P_n\}$ is necessarily tight. $\endgroup$ – John Dawkins Feb 5 '16 at 17:57
  • $\begingroup$ @JohnDawkins How can we use the fact that $\{ \mathbb{P}_n \}$ is tight??? $\endgroup$ – Richard Feb 5 '16 at 18:01

Fix $\varepsilon$. First use tightness to find a compact subset $K=K(\varepsilon)$ of $C[0,+\infty)$ such that $\mathbb P_n(K)\gt 1-\varepsilon$. Use the uniform convergence of $(f_n)_{n\geqslant 1}$ to $f$ in order to handle the integral of $f_n$ over $K$. Use the uniform bound to handle the integral of $f_n$ over the complement of $K$ (which has a measure $\mathbb P_n$ which does not exceed $\epsilon$).

| cite | improve this answer | |
  • $\begingroup$ My main problem now is that I can't bound the integral $\big| \int_K f_n \,d \mathbb{P}_n - \int_K f \, d \mathbb{P} \big|$. We somehow have to relate $\int_K f \,d \mathbb{P}_n$ to $\int_K f \,d \mathbb{P}$. $\endgroup$ – Richard Feb 6 '16 at 12:40
  • 1
    $\begingroup$ Write $\int f_n \,d\Bbb P_n-\int f \,d\Bbb P=\int(f_n-f)d\Bbb P_n+(\int fd\Bbb P_n-\int f\,d\Bbb P)$. The second term on the right converges to $0$ because of the assumed weak convergence. Deal with the first by breaking it up as $\int_K+\int_{K^c}$ and then proceeding as Davide Giraudo suggests. $\endgroup$ – John Dawkins Feb 6 '16 at 18:10

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.