10
$\begingroup$

Could any one tell me how to solve this one?

Given $f\in C[0,\infty)$ such that $f(x)\to 0$ as $x\to\infty$ we need to show that for any $\epsilon>0$ there is a polynomial $p$ such that $|f(x)-e^{-x}p(x)|<\epsilon \qquad \forall~ x\in[0,\infty)$

I just know the statement of Weierstrass Polynomial Approximation Theorem and that seems very far from the given problem, but somehow I feel I need to apply the theorem.

$\endgroup$
  • 5
    $\begingroup$ There is one rather complicated solution here. $\endgroup$ – Milind May 9 '13 at 17:37
  • 1
    $\begingroup$ :-o :-o :-o so big solution :-o $\endgroup$ – Marso May 9 '13 at 17:41
  • 5
    $\begingroup$ @milind I upvoted your comment, but I don't agree with the term complicated. The idea is that we can define $f(+\infty):=0$ and we transform continuously $[0,+\infty]$ into $[0,1]$ in order to get a compact interval. Then we approximate. A further approximation (which needs technical verifications) gives what we want. So the given solution is actually conceptually simple, and I would be surprised if a simpler solution would exist. $\endgroup$ – Davide Giraudo May 10 '13 at 9:47
  • $\begingroup$ @Davide Giraudo Hmm, thank you for the explanation. I confess that I did not understand the basic motivation behind the solution. Does the term "compact" here have a technical meaning? $\endgroup$ – Milind May 10 '13 at 9:52
  • $\begingroup$ @DavideGiraudo, superb explanation, Such a big proof needs this type of explanations. thanks .. $\endgroup$ – ROBINSON May 13 '13 at 11:43
1
$\begingroup$

It's not a "solution" (proof?) but I would like to give some details to your question.

$$ \vert f(x)-e^{-x}p(x)\vert=e^{-x}\vert g(x)-p(x)\vert, $$ where $\lim_{x\to\infty}e^{-x}g(x)=0$. Your question is about the density of polynomials in the weighted space (weight function is $e^{-x}$) on the half-line. More info can be found Stone-Weierstrass theorem (locally compact version). The literature of weighted approximation in $L_p$ spaces is huge.

$\endgroup$
  • $\begingroup$ appreciate your answer, thanks $\endgroup$ – Betty Mock Jan 10 '14 at 4:44

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.