Take the 2-minute tour ×
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.

I am currently studying Fourier Analysis on my own and have just been started to look at the Schwartz Space of rapidly decaying functions. One example of such functions is given in the notes that I use by \begin{equation} f(x) = e^{-x^2} \end{equation}

I tried to come up with an argument that shows this is actually true. Here is what I did:

for any $m$ we have $f^{(m)}(x) = p(x)e^{-x^2}$ where $p(x) = ax^2 + bx + c$ is a second order polynomial. Hence the function \begin{equation} h(x) = x^kf^{(m)} = x^kp(x)e^{-x^2} \end{equation} is smooth. Furthermore, by repeated use of L'Hopitals Rule we have \begin{equation} \lim_{x \to \pm \infty} h(x) = 0 \end{equation} Hence $h$ must be bounded by some constant $c_{k,m}$, as required.

Now, my question is whether the argument is valid and whether there is a more standard way of showing this. I hope the question does not sound too dumb! I am a newbie at this and cannot fall back on feedback via homework, so was hoping to get some comment here on my attempt. Thanks!

share|improve this question
add comment

1 Answer 1

up vote 3 down vote accepted

You have the idea, but the problem is that the polynomial $p$ depends on $m$ (and also its degree). We can define $p_0(x)=1$ and $$\frac d{dx}(p_m(x)e^{-x^2})=p_m'(x)e^{-x^2}-2xp_m(x)e^{-x^2}=e^{-x^2}(p_m'(x)-2xp_m(x)),$$ so we have to define $p_{m+1}(x)=p_m'(x)-2xp_m(x)$. We have $\deg p_{m+1}=1+\deg p_m$ so $\deg p_m=m$ and for $k,m$ integers: $$|x^kf^{(m)}(x)|=|x^kp_m(x)e^{-x^2}|\leq \frac{|x^kp_m(x)|}{\sum_{l=0}^{m+k+1}\frac{x^l}{l!}},$$ which is bounded since $\deg (x^kp_m)=k+m<m+k+1$.

share|improve this answer
    
ah so I didn't analyse the behaviour of the derivatives of $e^{-x^2}$ in enough detail, thanks alot for the answser that helped a great deal! –  harlekin Jan 30 '12 at 21:48
    
In fact the derivatives of $e^{-x^2}$ are linked to Hermite polynomials. –  Davide Giraudo Jan 30 '12 at 21:50
    
that's interesting, I'll check them out! –  harlekin Jan 30 '12 at 22:07
add comment

Your Answer

 
discard

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.