Convergence of Geometric Series .... Sohrab, Proposition 2.3.8 .... ....

Click For Summary
SUMMARY

The discussion centers on the proof of Proposition 2.3.8 from Houshang H. Sohrab's "Basic Real Analysis" (Second Edition), specifically addressing the convergence of geometric series when the absolute value of the ratio \( |r| < 1 \). Participants clarify that by substituting \( b = |r| \), one can conclude \( \lim (|r|^n) = 0 \), which leads to \( \lim (r^{n+1}) = 0 \) through established limit definitions. The use of Cauchy sequences is also suggested as a robust method to demonstrate convergence in this context.

PREREQUISITES
  • Understanding of limits in real analysis
  • Familiarity with geometric series and their convergence criteria
  • Knowledge of Cauchy sequences and their properties
  • Proficiency in mathematical notation and proofs
NEXT STEPS
  • Study the concept of Cauchy sequences in detail
  • Review the proof techniques for geometric series convergence
  • Explore the implications of Exercise 2.2.7 in Sohrab's book
  • Learn about the properties of limits and their applications in real analysis
USEFUL FOR

Students of real analysis, mathematicians focusing on series convergence, and educators teaching advanced calculus concepts will benefit from this discussion.

Math Amateur
Gold Member
MHB
Messages
3,920
Reaction score
48
I am reading Houshang H. Sohrab's book: "Basic Real Analysis" (Second Edition).

I am focused on Chapter 2: Sequences and Series of Real Numbers ... ...

I need help with an aspect of the proof of Proposition 2.3.8 ...

Proposition 2.3.8 and its proof read as follows:
View attachment 9048
In the above proof by Sohrab we read the following:

" ... ... Now, if $$\mid r \mid \lt 1$$ then (by Exercise 2.2.7) $$\lim (r^{ n + 1}) = 0$$ and $$( \ast )$$ follows at once ... ... "
Now it appears that Sohrab is referring to Part 2 of Exercise 2.2.7 (see below) that states:

" ... ... if $$0 \lt b \lt 1$$, then $$\lim ( b^n ) = 0$$ ... ... "Now my problem is we are not given $$0 \lt r \lt 1$$ ... but instead we are given $$\mid r \mid \lt 1$$ ... .. so my question is ... what do we do here ...?
Perhaps we put $$b = \mid r \mid$$ ... but this allows us to conclude $$\lim ( \mid r \mid^n ) = 0$$ ...... can we conclude from here that $$\lim (r^{ n + 1}) = 0$$ ...... by arguing that $$\lim ( \mid r \mid^n ) = 0 = \lim ( \mid r^n \mid ) = \lim ( \mid r^{ n +1} \mid )$$... and further by arguing that if $$\lim \mid r^{ n +1} \mid = 0$$ ...... then ...... $$\lim r^{ n +1} = 0$$ ...
Can someone please comment on my reasoning to resolve the problem I had with Sohrab's assertion that appeal to Exercise 2.2.7 would lead to $$( \ast )$$ following at once ...
Help will be appreciated ...

Peter
========================================================================================
The above post refers to Exercise 2.2.7 ... so I am providing text of the same ... as follows:View attachment 9049
Hope that helps ...

Peter
 

Attachments

  • Sohrab - Proposition 2.3.8 ... .png
    Sohrab - Proposition 2.3.8 ... .png
    14.6 KB · Views: 134
  • Sohrab - Exercise 2.2.7 ... .png
    Sohrab - Exercise 2.2.7 ... .png
    10.9 KB · Views: 149
Physics news on Phys.org
A nice way to do this would be with cauchy sequences, but I don't know if you've gotten to them yet. The argument would be

$m \gt n \geq N$$\big \vert s_n - s_m\big \vert $
$ = \big \vert\big(\frac{a}{1-r}\big)\big(1 -r^{n+1} -(1 -r^{m+1})\big)\big \vert $
$ = \big \vert\big(\frac{a}{1-r}\big)\big(- r^{n+1} + r^{m+1}\big)\big \vert $
$ = \big \vert\frac{a}{1-r}\big \vert \big \vert r\big \vert^{n+1}\cdot \big \vert r^{m-n} - 1 \big \vert $
$ \lt 2 \cdot \big \vert\frac{a}{1-r}\big \vert \big \vert r\big \vert^{n+1} $

where the inequality follows from triangle inequality

select $\frac{\epsilon}{2}$
i.e. based on earlier work with the non-negative case, you know
$\big(\big \vert\frac{a}{1-r}\big \vert \big \vert r\big \vert^{n+1}\big)$ tends to zero... so re-use this result
- - - -
as far as your suggested approach, I think it works.

You can also observe that
$\big \vert r\big \vert^n =\big \vert r^n \big \vert = \big \vert r^n - L \big \vert$
with $L =0$

since you know from that prior exercise 2.2.7 that for any $\epsilon \gt 0$
$\big \vert r\big \vert^n \lt \epsilon$
by selecting large enough $N$, then you also know by selecting that same $N$ that
$\big \vert r^n - L \big \vert \lt \epsilon$
which is your definition of a limit
 
steep said:
A nice way to do this would be with cauchy sequences, but I don't know if you've gotten to them yet. The argument would be

$m \gt n \geq N$$\big \vert s_n - s_m\big \vert $
$ = \big \vert\big(\frac{a}{1-r}\big)\big(1 -r^{n+1} -(1 -r^{m+1})\big)\big \vert $
$ = \big \vert\big(\frac{a}{1-r}\big)\big(- r^{n+1} + r^{m+1}\big)\big \vert $
$ = \big \vert\frac{a}{1-r}\big \vert \big \vert r\big \vert^{n+1}\cdot \big \vert r^{m-n} - 1 \big \vert $
$ \lt 2 \cdot \big \vert\frac{a}{1-r}\big \vert \big \vert r\big \vert^{n+1} $

where the inequality follows from triangle inequality

select $\frac{\epsilon}{2}$
i.e. based on earlier work with the non-negative case, you know
$\big(\big \vert\frac{a}{1-r}\big \vert \big \vert r\big \vert^{n+1}\big)$ tends to zero... so re-use this result
- - - -
as far as your suggested approach, I think it works.

You can also observe that
$\big \vert r\big \vert^n =\big \vert r^n \big \vert = \big \vert r^n - L \big \vert$
with $L =0$

since you know from that prior exercise 2.2.7 that for any $\epsilon \gt 0$
$\big \vert r\big \vert^n \lt \epsilon$
by selecting large enough $N$, then you also know by selecting that same $N$ that
$\big \vert r^n - L \big \vert \lt \epsilon$
which is your definition of a limit
Thanks steep ...

Most grateful for your help ...

Peter
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
2K
Replies
4
Views
2K
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
3
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
2
Views
2K