Apostol's Calculus Vol.1 10.20 #20

  • Thread starter Thread starter Masaki
  • Start date Start date
  • Tags Tags
    Calculus
Click For Summary
SUMMARY

The series \(\sum_{n=1}^{\infty} (-1)^n \left(\frac{\pi}{2}-\text{arctan}({\log{n}})\right)\) is conditionally convergent as established by Leibniz's rule. However, it does not converge absolutely, which can be demonstrated using the limit comparison test with \(a_n = \arctan(1/\log{n})\) and \(b_n = 1/\log{n}\). The limit \(\lim_{n\rightarrow\infty} \frac{a_n}{b_n} = 1\) confirms that since \(\sum b_n\) diverges, \(\sum a_n\) must also diverge, thus proving the lack of absolute convergence.

PREREQUISITES
  • Understanding of series convergence tests, specifically Leibniz's rule and the limit comparison test.
  • Familiarity with the properties of the arctangent function and logarithmic functions.
  • Knowledge of l'Hôpital's rule for evaluating limits.
  • Basic calculus concepts related to infinite series and convergence.
NEXT STEPS
  • Review the application of the limit comparison test in series convergence.
  • Study the properties and behavior of the arctangent function in relation to logarithmic growth.
  • Explore advanced convergence tests such as the ratio test and root test.
  • Practice problems involving conditional and absolute convergence of alternating series.
USEFUL FOR

Students studying calculus, particularly those focusing on series convergence, as well as educators and tutors seeking to enhance their understanding of convergence tests in mathematical analysis.

Masaki
Messages
8
Reaction score
0

Homework Statement


Determine (absolute/conditional) convergence or divergence of the series
\displaystyle\sum_{n=1}^{\infty} (-1)^n \left(\frac{\pi}{2}-\text{arctan}({\log{n}})\right)​

The Attempt at a Solution


It is easy to see that the series is (conditionally) convergent by Leibniz's rule, but I am not sure how to deny the absolute convergence of the series. To use the comparison test, I must show, for example,
f(x) = \displaystyle\left(\frac{\pi}{2}-\text{arctan}(\log{x})\right) - \frac{1}{x} > 0​
for all x \geq 1, which seems quite difficult (in fact, I have no idea how to prove it). Is there any other, better way to show that the series is not absolutely convergent?
 
Physics news on Phys.org
Keep in mind
\arctan(1/x)=sgn(x)\pi/2 - \arctan(x)
 
susskind_leon said:
Keep in mind
\arctan(1/x)=sgn(x)\pi/2 - \arctan(x)

Thanks for the reply! I think the problem now becomes manageable.
Let a_n = \arctan(1/\log{n}) and b_n = 1/\log{n}. Then, we have, with using l'Hôpital's rule,
\displaystyle \lim_{n\rightarrow\infty} \frac{a_n}{b_n}<br /> = \lim_{n\rightarrow\infty} \frac{a_n&#039;}{b_n&#039;}<br /> = \lim_{n\rightarrow\infty} \frac{ \frac{1}{1 + 1/(\log{n})^2}\cdot (1/\log{n})&#039; }{(1/\log{n})&#039;}<br /> = \lim_{n\rightarrow\infty} \frac{1}{1 + 1/(\log{n})^2} = 1.​
Since \sum b_n = \sum (1/\log{n}) diverges, by the limit comparison test, \sum a_n also diverges.
 

Similar threads

  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
5
Views
2K
Replies
7
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 13 ·
Replies
13
Views
2K