Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

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

On a metric space $X$, Did said:

the space $M_1$ of probability measures is included in the dual of the space $C_b$ of bounded continuous functions.

I was wondering what is the representation theorem for the dual of $C_b(X)$, which is supposed to have $M_1$ as its subset?

I saw in Wikipedia, only Riesz representation theorems for $C_0$ and $C_c$, not for $C_b$.

Thanks and regards!

share|cite|improve this question
No representation theorem needed. Just integrate with respect to $\mu \in M_1$ to get a continuous linear functional on $C_b(X)$. – Martin Feb 21 '13 at 3:16
@Martin: Thanks! (1) In that scenario, is $C_b(X)$ endowed with the the supremum norm? (2) Can the continuous dual of $C_b(X)$ be represented as a set of measures which contains $M_1$? – Tim Feb 21 '13 at 3:22
(1) Yes. (2) No. – Martin Feb 21 '13 at 3:23
You edited your question (2)... For the new (2): $C_b(X) = C(\beta X)$ and apply the Riesz representation theorem for $\beta X$, so it is a set of measures, but not measures on $X$. – Martin Feb 21 '13 at 3:28
I'm not a person to say what Did can and what not. Anyway, he already said that, and that's true. Even more, $\mathcal M_1$ is a linear subspace of a continuous dual of $\mathrm b\mathfrak B_X$ - the Banach space of bounded measurable function for any measurable space $(X,\mathfrak B_X)$ endowed with a $\sup$-norm. To state that $\mathcal A$ is a subset of a continuous dual of $\mathcal B$ you only have to check that any element of $\mathcal A$ acts continuously on $\mathcal B$, the actual dual $\mathcal B^*$ does not have to be computed. – Ilya Feb 22 '13 at 17:13
up vote 2 down vote accepted

For the sake of having an answer to the question, Martin's comment contains the key point. We have an isomorphism $C_b(X) \cong C(\beta X)$ where $\beta X$ is the Stone-Čech compactification (defined for arbitrary topological spaces; some authors put extra conditions on $X$ but these are just the conditions required for the natural map $X \to \beta X$ to be an embedding), so by the ordinary Riesz representation theorem describes the dual of $C_b(X)$ in terms of measures on $\beta X$.

share|cite|improve this answer
Thanks! But still not the same as what Did say. He said the probability measures on $X$ instead of $\beta X$ is a subset of dual of $C_b(X)$. For that purpoe, shall the measures on $βX$ be restricted to $X$ (we can do this because X is metric space and thus Tychonoff, the mapping from X to its image in βX is injective and homeomorphism)? – Tim Feb 22 '13 at 21:30
As Martin has already said, you don't need a representation theorem to see this. Just integrate measures on $X$ against functions in $C_b(X)$. – Qiaochu Yuan Feb 22 '13 at 21:31

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.