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

I am trying to show that for $t$ in $(0, T]$ for some $T > 0$ and $t |y| \leqslant C$ for some $C > 0$ that

$$\tag{1} |x|^M \exp \left (-\frac{D}{1 - \rm e^{-2 t^2}} {|e^{-t^2} x - y|^2}{} \right )$$ is uniformly bounded in $x, y$ (in $\mathbf R^d$) and $t$. The implied constant may depend on $D > 0$ and integers $M$. It is also possible to choose $D > 0$ as long as it does not depend on any of the quantities.

Alright! So let us set $z = t y$ then $|z| \leqslant C$. Right. So we can rewrite $(1)$ as $$\tag{2} |x|^M \exp \left (-\frac{D}{t (1 - \rm e^{-2 t^2})} {|e^{-t^2} t x - z|^2}{} \right ).$$

We could also note that we can just consider real numbers $x$ and $y$ and afterwards take products. That might simplify the analysis a bit. So we "deduce" from $(2)$ that we should find a bound for (abusing $M$ a bit) $$\tag{3} x^{2M} \exp \left (-\frac{D}{t (1 - \rm e^{-2 t^2})} {(e^{-t^2} t x - z)^2}{} \right ).$$

This is where misery starts. I have started expanding the square and then using cases but then stuff starts depending on each other. Ugly.

Does someone have any idea? A suggestion perhaps?

share|cite|improve this question
up vote 3 down vote accepted

For $t\in(0,T\;]$, $$ \frac{D}{1-e^{-2t^2}}\ge\frac{D}{1-e^{-2T^2}}\tag{1} $$ If $|x|\ge2(C/t)e^{T^2}$, then $e^{-t^2}|x|>2|y|$, and therefore, $$ \left|e^{-t^2}x-y\right|^2\ge\frac{1}{2}e^{-t^2}|x|\ge\frac{1}{2}e^{-T^2}|x|\tag{2} $$ Thus, $$ \frac{D}{1-e^{-2t^2}}\left|e^{-t^2}x-y\right|^2\ge\left(\frac{D/2\;\;e^{-T^2}}{1-e^{-2T^2}}\right)|x|\tag{3} $$ So that $$ |x|^M\exp\left(-\frac{D}{1-e^{-2t^2}}\left|e^{-t^2}x-y\right|^2\right)\le|x|^M\exp\left(-\left(\frac{D/2\;\;e^{-T^2}}{1-e^{-2T^2}}\right)|x|\right)\tag{4} $$ which is obviously bounded.

However, let $y=e^{-t^2}x$ and $t=C/|x|$. Then, for $|x|>C/T$, we have $t|y|<C$, $t<T$, and $$ |x|^M\exp\left(-\frac{D}{1-e^{-2t^2}}\left|e^{-t^2}x-y\right|^2\right)=|x|^M\tag{5} $$ which can get arbitrarily large.

share|cite|improve this answer

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.