Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Suppose $\{\mu_{n}\}$ is a sequence of probability measure on $\mathbb{R}$ such that $\mu_{n}(A)=\delta_{\alpha_{n}}(A)=\begin{cases} 1, & \alpha_{n}\in A\\ 0, & \alpha_{n}\notin A \end{cases}.$ Suppose $\delta_{\alpha_{n}}\overset{w}{\rightarrow}\mu$. Show that $\exists\alpha$ such that $\mu=\delta_{\alpha}$ and $\alpha_{n}\rightarrow\alpha$ .

Proof: $F_{n}(x)=\mu_{n}(-\infty,x]=\chi_{[\alpha_{n},\infty)}$


I don't understand the following step:

$\mu_{n}\overset{w}{\rightarrow}\mu\Rightarrow F_{n}(x)\rightarrow F(x)$ on a dense subset$\Rightarrow F(x)=\mu(-\infty,x]=\chi_{[\alpha,\infty)}$ where $\alpha_n\rightarrow\alpha$.

Question: (1) Do I need to show the specific relationship between $\alpha_n$ and $\alpha$? For example, take the sup or inf. The above proof seems to naturally assumes that $\lim \alpha_n$ exists. (2) Does the dense set have to be any specific dense set?

Assume that we can change the question. Assume $\delta_{\alpha_{n}}\overset{w}{\rightarrow}\delta_{\alpha}$. Then, I can easily prove $\alpha_{n}\rightarrow\alpha$. So I guess the real problem for me is $\mu=\delta_{\alpha}$.

Thank you!

share|cite|improve this question
up vote 2 down vote accepted

The functions $F_n$ have the special property that $F_n(x)$ can only be zero or one, for every $x$. For every $x$ in the dense set where $F_n(x)\to F(x)$, the same is true for $F(x)$, i.e., either $F(x)=0$ or $F(x)=1$. Since $F$ is right continuous, we have $F(x)=0$ or $F(x)=1$ for every $x\in \mathbb{R}$.

The function $F$ is non-decreasing and non-constant, and has $F(x)\in\{0,1\}$ for every $x$. The only such functions are $\chi_{[\alpha,\infty)}$ for $\alpha\in\mathbb{R}$.

Added: (1) You do not need to show a specific relationship. The proof does not assume that $\lim_n \alpha_n$ exists. As in my solution, you first show that $\alpha$ exists, and then prove that $\alpha_n\to\alpha$. You yourself said that this is easy.

(2) Any dense set will do.

share|cite|improve this answer
Thank you, Byron. I edited the question to make it more specific. Could you please further clarify it? Thanks. – user16859 Apr 9 '12 at 1:38

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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