2
$\begingroup$

I am trying to understand the proof behind Levy's theorem. The statement is as follows

Theorem: Let $(X_n)_{n\in \mathbb N}$ be a family of random variables, $(\mu_n)_{n\in \mathbb N}$ their distributions and $(\varphi_n)_{n\in \mathbb N}$ their characteristic functions.

Suppose that for every $t \in \mathbb R$ $$\lim_{n\to \infty}\varphi_n(t)= \varphi(t)$$ and that $\varphi$ is continuous in $0$. Then there exists a distribution $\mu$ such that $\varphi$ is $\mu$'s characteristic function and $(\mu_n)_{n \in \mathbb N}$ converges weakly to $\mu$.

The proof that I have been provided with shows that the family $(\mu_n)_{n\in \mathbb N}$ is tight through an integral argument.

From my (limited) knowledge in probability, Prokhorov's theorem states that if the family $(\mu_n)_{n\in \mathbb N}$ is tight then there exists a distribution $\mu$ and a subsequence of the given family $(\mu_{\kappa_n})_{n\in \mathbb N}$ that converges weakly to $\mu$, but there is no general result about the convergence of $(\mu_n)_{n \in \mathbb N}$.

Supposing that we have shown that $(\mu_n)_{n\in \mathbb N}$ is tight, we can conclude that there exists a sequence $\kappa_n$ such that $$\lim_{n\to \infty}\varphi_{\kappa_n}(t)=\varphi(t)$$ but

How can we conclude that $(\mu_n)_{n \in \mathbb N}$ converges weakly to $\mu$?

$\endgroup$
1
$\begingroup$

Here are the pieces you are missing, I think:

  1. Prokhorov's theorem actually says that $\textit{every}$ subsequence of your $(\mu_n)_{n \in \mathbb{N}}$ has a sub-subsequence converging in the weak topology to some probability measure.
  2. By your condition on the sequence $(\phi_n)_{n \in \mathbb{N}}$ converging to $\phi$, every one of the sub-subsequences of $(\mu_n)_{n \in \mathbb{N}}$ must converge to the measure $\mu$ whose characteristic function is $\phi$.
  3. In a Polish space, the topology of weak convergence is metrizable.

That means limits are unique, and the sequence $(\mu_n)_{n \in \mathbb{N}}$ converges weakly to a probability measure $\mu$ if and only if each subsequence has a further subsequence converging weakly to that measure --- true because of steps 1, 2.

$\endgroup$

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.