How to Solve Improper Integral: Tips & Tricks

AI Thread Summary
The discussion focuses on solving the improper integral ∫_{-\infty}^{\infty}e^{-(ax^2+b/x^2)}dx, which equals √(π/a)e^{-2√(ab)}. Participants share techniques, particularly differentiating under the integral sign and changing variables, to derive the solution. A specific method involves defining the integral I, finding its derivative I' with respect to √b, and applying a variable substitution. Additional resources, including papers that utilize this integral in probability calculations related to Brownian motion, are shared for further reference. The conversation highlights the importance of practice in applying advanced calculus techniques.
techmologist
Messages
305
Reaction score
12
How do you get this:

\int_{-\infty}^{\infty}e^{-(ax^2+b/x^2)}dx = \sqrt{\frac{\pi}{a}}e^{-2\sqrt{ab}}

I've been trying all the tricks I know, like differentiating under the integral sign and whatnot, but I can't get it. Thanks.
 
Mathematics news on Phys.org
In which course did you run into this monster?
 
Not in any course. I came across it in a paper where they calculate the probability that a Brownian motion X(t) crosses the line at+b at some point during the interval (0, T).
 
Send a link of the paper?
 
1)Let I be your integral
2)Find I' by differentiating under the integral sign with respect to sqrt(b)
3)Effect the change of variable u=sqrt(b/a)/x (break up the interval at 0)
4)Notice I=-2sqrt(a)I'
5)Conclude I=I0 exp(-2sqrt(ab))
6)Compute I0

sqrt() denotes the square root
I' is the derivative of I with respect to sqrt(b)
I0 is your integral when b=0
See this thread
Feynman's Calculus
 
lurflurf said:
1)Let I be your integral
2)Find I' by differentiating under the integral sign with respect to sqrt(b)
3)Effect the change of variable u=sqrt(b/a)/x (break up the interval at 0)
4)Notice I=-2sqrt(a)I'
5)Conclude I=I0 exp(-2sqrt(ab))
6)Compute I0

sqrt() denotes the square root
I' is the derivative of I with respect to sqrt(b)
I0 is your integral when b=0
See this thread
Feynman's Calculus

Thanks, lurflurf! That's exactly what I needed. Well done. That's an excellent thread you linked to, also. Surely, You're Joking is where I first learned about differentiating under the integral sign. I apparently need more practice at applying it, though.

Coriolis314 said:
Send a link of the paper?

Sure. Here is the paper where the probability is calculated (near the end) using Laplace transforms:

http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.pjm/1102911625

Here is a paper that makes use of the result. Scroll down to where it says "Kelly Criterion"--that's it. In the section on probability of reaching a goal within a certain time.

http://edwardothorp.com/id10.html
 
Last edited:
Seemingly by some mathematical coincidence, a hexagon of sides 2,2,7,7, 11, and 11 can be inscribed in a circle of radius 7. The other day I saw a math problem on line, which they said came from a Polish Olympiad, where you compute the length x of the 3rd side which is the same as the radius, so that the sides of length 2,x, and 11 are inscribed on the arc of a semi-circle. The law of cosines applied twice gives the answer for x of exactly 7, but the arithmetic is so complex that the...
Back
Top