Series expansion of an integral at infinity

Click For Summary
SUMMARY

The discussion centers on the series expansion of integrals at infinity, specifically examining the expression involving two divergent integrals: \(\left[\int_0^{\infty} \frac{1}{x^n}e^{1/x}\, dx \right] \left[\int_{-\infty}^{+\infty}e^{u^2} \cos u\, du \right]^{-1}\). The user queries whether the ratio converges to a finite number, particularly when \(n=2\), suggesting that both integrals have a series expansion at infinity represented by \(e^{x^2}\). The conversation highlights the importance of defining functions for the integrals to facilitate a Taylor expansion and discusses the implications of the radius of convergence.

PREREQUISITES
  • Understanding of integral calculus, particularly improper integrals.
  • Familiarity with series expansions and Taylor series.
  • Knowledge of the behavior of exponential functions at infinity.
  • Experience with mathematical software tools like Wolfram Alpha for computational analysis.
NEXT STEPS
  • Explore the concept of series expansion for integrals, focusing on techniques for handling divergent integrals.
  • Study the properties of the exponential function and its series expansion at infinity.
  • Learn about Taylor series and their applications in approximating functions near specific points.
  • Investigate the implications of the radius of convergence in the context of series expansions of integrals.
USEFUL FOR

Mathematicians, physicists, and students engaged in advanced calculus, particularly those interested in series expansions and the behavior of integrals at infinity.

Irid
Messages
207
Reaction score
1
Hello,
I'm fiddling with Wolfram Alpha and I can't find a definition of what do they mean by the "Series expansion of the integral at x -> inf". In particular, I have two divergent integrals and I am wondering whether their ratio is some finite number. Here it is:

\left[\int_0^{\infty} \frac{1}{x^n}e^{1/x}\, dx \right] \left[\int_{-\infty}^{+\infty}e^{u^2} \cos u\, du \right]^{-1}

where n is a parameter. Based on wolfram's suggestion, I think that if n=2, the above expression converges to something meaningful, since both integrals apparently have the series expansion at infinity as exp(x^2).
 
Physics news on Phys.org
I do not see any functions there with ##n## specified. The integrals are definite integrals, i.e. numbers or infinite. Hence the quotient is another number or undefined.

If you manage to make functions out of them, then you can of course write down a Taylor expansion. Radius of convergence is then a different question.
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 6 ·
Replies
6
Views
4K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 17 ·
Replies
17
Views
5K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 19 ·
Replies
19
Views
4K
Replies
3
Views
2K