My Question reads:

Recall that the Alternating Series Test says that if $a_n\to 0$ and $a_n\geq\ a_{n+1}$ for all $n\geq\ 1$, then the alternating series $\sum\limits_{n=1}^{\infty} (–1)^{n+1}a_n$ converges; i.e., the sequence of partial sums $s_n= a_1–a_2+ a_3–... ± a_n$ converges.

Consider the subsequences ($s_{2n}$) and ($s_{2n-1}$), and show how the Monotone Convergence Theorem proves the Alternating Series Test.

My work:

Since we need to use the MCT, I was thinking of first showing that the subsequences ($s_{2n}$) and ($s_{2n-1}$) are bounded and monotone. So for the even subsequences I wanted to show they are increasing & bounded above while the odds are decreasing & bounded below.

However, I am having some trouble with this. So, I am not sure if I am proving by induction on $n$ that $s_{2n}\leq\ s_{2n-1}$ or instead that $s_{2n}\leq\ s_1$ and $s_2\leq\ s_{2n-1}$.

Also, to show the evens are increasing, we would need to show $s_{2n+2}-s_{2n}\geq\ 0$, but would I need to first show the other inductions? I am not sure how to show rigorously that the evens are increasing while the odds are decreasing.


Does this work.

I will show by induction on $n$ that $s_{2n}\leq\ s_1$.

For the base case of $n=1$ we have that $a_1-a_2<a_1$ so it is true. Then we assume $s_{2n}\leq\ s_1$.

Then, $s_{2(n+1)}=s_{2n+2}= s_1-a_{2n+2}\leq\ s_1-a_{2n+2}<a_1$. So this shows bounded above.

How can I show increasing in this case?

How can I show that $s_{2n+2}-s_{2n}\geq\ 0$?

I need help with the odd sequences part of the proof. I am not too sure how to show they are bounded below by $s_2$ because I tried induction and it did not work out.


Using induction you can easily prove that $s_{2k+1} \geq s_{2k+3} $ for any $k \geq 0$ and that it also bounded below by zero. Therefore by the Monotone Convergence Theorem the sequence $(s_{2n+1})$ has a limit denoted $l_1 \geq 0$

On the other hand, using an induction argument you can show that $s_{2k} \leq s_{2k+2}$ for any $k \geq 1$, so the sequence $(s_{2n})$ has a limit $l_2$ (Careful as $l_2$ might be $\infty$ apriori).

Finally, using the fact that $a_n\to 0$, conclude that $l_1=l_2$.

| cite | improve this answer | |
  • $\begingroup$ I am not sure how you are doing that first induction. Aren’t we showing that the odd sequences are bounded below by say $s_2$ and the evens are bounded above by say $s_1$? $\endgroup$ – Sam Oct 22 '17 at 17:23
  • $\begingroup$ Then we also need to show that the evens are increasing and the odds are decreasing? $\endgroup$ – Sam Oct 22 '17 at 17:24
  • $\begingroup$ I do not get how you would do the second induction you are suggesting. $\endgroup$ – Sam Oct 22 '17 at 18:58
  • $\begingroup$ You have that $s_{2k+2}=s_{2k}+a_{2k}-a_{2k+1}$, and then use $a_{2k} \geq a_{2k+1}$ $\endgroup$ – C.Niculescu Oct 22 '17 at 22:55
  • $\begingroup$ Right, I got it, thank you. Do you know how we can split up $s_{2n-1}$? $\endgroup$ – Sam Oct 23 '17 at 0:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.