I'm trying to show this by using Stone-Weierstrass Theorem.

For clarity, let me state definitions and the theorem first :)

Definition: Uniform metric

Let $\overline{d}$ be the standard bounded metric induced by the standard metric $d$ on $\mathbb{C}$.

Define $\overline{\rho}(f,g)\triangleq \sup_{x\in X} \overline{d}(f(x),g(x))$.

Call $\overline{\rho}$ "the uniform metric on $\mathbb{C}^X$ induced by $d$".

(This is the definition given in Munkres-Topology)


Stone-Weierstrass Theorem (complex version)

Let $X$ be a locally compact space.

Let $C_0(X,\mathbb{C})$ be equipped with the topology induced by the uniform metric $\overline{\rho}$.

Let $\mathscr{A}$ be a subalgebra of $C_0(X,\mathbb{C})$

If $\mathscr{A}$ is self-adjoint, separates points and vanished nowhere, then $\mathscr{A}$ is dense in $C_0(X,\mathbb{C})$.

Now, let me illustrate what exactly my question is.

Set $X$ a compact subset of $\mathbb{C}$.

Since $X$ is compact, $C_0(X,\mathbb{C})=C(X,\mathbb{C})$

I want to know whether the set $P\triangleq \{f\in \mathbb{C}^X:f \text{ is a polynomial function}\}$ is dense in $C(X,\mathbb{C})$.

To apply the theorem, I tried to show that $P$ is self-adjoint, but I couldn't. Now, I don't really think that $P$ is self-adjoint, but if so, how do I prove this?

Even if $P$ is not self-adjoint, I believe it's true that $P$ is dense in $C(X,\mathbb{C})$. How do I show this?


Depends of the interpretation of "polynomial function". Let be $z=x+iy$. If your polynomials are elements of $\Bbb C[x,y]$, then the self-adjointness condition is true. But $\Bbb C[z]$ isn't self-adjoint.


Look at the case of $X = D^2$, the unit disc in $\mathbb{C}$. A uniform limit of polynomials on this domain is analytic, so the uniform closure is contained in the analytic functions on $D^2$.


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.