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

This post can be thought of as the prototype proof and the motivation for the question posted here Using a laplace type expansion to get bounds on an integral arising in the study of Brownian motion

Let $ 0 < r < 1$, fix $x > 1$ and consider the integral

$$ I_{1}(x) = \int_{1}^{\infty} \exp\left( - \frac{x^2}{2y^{2r}} - \frac{y^2}{2}\right) \frac{dy}{y^r}.$$

The result we have been trying to prove below can be thought of as an analysis of localizing around the maximum of the function $f(y) = \frac{x^2}{2y^{2r}} + \frac{y^2}{2}$ in the exponential of the integrand and then applying appropriate estimates along with integration by parts. Roughly speaking I am having trouble with the analysis on one side of the maximum.

Let $I_1(x) = \int_{1}^{\infty} \exp( -\frac{x^2}{2y^{2r}} - \frac{y^2}{2} ) \frac{dy}{y^r}$ and fix a constant $c^* = r^{\frac{1}{2r+2}} $ that is $c^* = argmin ( d(c) )$ where $d(c) = \frac{1}{2} (c^{-2r} + c^2)$ and let $x^* = x^{\frac{1}{1+r}}$. Then we break up the integral as

$$I_1(x): = I_{11}(x) + I_{12}(x) = \int_{1}^{c^* x^*} \exp( -\frac{x^2}{2y^{2r}} - \frac{y^2}{2} ) \frac{dy}{y^r} + \int_{c^*x^*}^{\infty} \exp( -\frac{x^2}{2y^{2r}} - \frac{y^2}{2} ) \frac{dy}{y^r} $$

The particular trouble I am having arises with estimating $I_{11}$.

Using integration by parts on $I_{11}(x)$ we set $f(y) = \frac{x^2}{2y^{2r}} + \frac{y^2}{2}$ and write $$ I_{11}(x) = \int_{1}^{c^* x^*} \exp( -\frac{x^2}{2y^{2r}} - \frac{y^2}{2} ) \frac{dy}{y^r} = \int_{1}^{c^* x^*} \frac{y^{-r}}{f'(y)} f'(y) e^{-f(y)} dy$$

Let $v = \frac{1}{y^r f'(y)}, du = f'(y) e^{-f(y)}dy$ so that $u = - e^{-f(y)}$ and $dv = \frac{-r y^{-r-1}f'(y) -y^{r} f''(y) }{(f'(y))^2}$ and write $$dv = \frac{-r y^{-r-1}}{f'(y)} dy - \frac{y^{-r} f''(y)}{(f'(y))^2}dy := v_1(y)dy + v_2(y)dy$$

Applying integration by parts to $I_{11}(x)$ we get

$$ \int_{1}^{c^* x^*} \frac{y^{-r}}{f'(y)} f'(y) e^{-f(y)} dy = \left[- \frac{y^{-r}}{f'(y)} exp(-f(y)) \right]_{1}^{c^* x^*} + \int_{1}^{c^* x^*} e^{-f(y)} (v_1(y) +v_2(y)) dy $$

1) Using this argument is it possible to show $I_{11}(x) \leq k(r,x)\exp(- x^{\frac{2}{1+r}} d(c^*))$ where $k(r,x)$ is a rational function in $x$?

The issue seems to be for us finding a way to remove the singularities around $v_1$ and $v_2$ near $c^* x^*$.

share|cite|improve this question

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.