Math 222 Chapter 11 Review 05/01/19. [p.

1]

1. (a) What is a convergent sequence?

Solution. {an }∞
n=1 is convergent if limn→∞ an exists.

(b) What is a convergent series?

P∞
Solution. n=1 an is convergent if the limit of partial sums limn→∞ (a1 + a2 + . . . + an ) exists.

(c) What does lim an = 3 mean?

n→∞

Solution. It means that for any error  > 0, the sequence an is eventually contained in the
interval (3 − , 3 + ). By taking n large enough, we can guarantee that the error between 3 and
an is arbitrarily small.

∞
P
(d) What does an = 3 mean?
n=1

Solution. Similar to the one above, it means that by taking k arbitrarily large, we can make
Pk
the error 3 − n=1 an arbitrarily small.

∞
X
2. Suppose an = 3 and sn is the nth partial sum of the series. What is lim an ? What is lim sn ?
n→∞ n→∞
n=1

Solution. Because the series is convergent, limn→∞ an = 0. And limn→∞ sn = 3 by definition of

series convergence.

True or false (explain)

∞
P
3. If lim an = 0, then an converges.
n→∞ n=0

P∞ 1
Solution. False. E.g. n=1 n .

∞
X
4. If cn xn diverges when x = 6, then it diverges when x = 10.
n=0

Solution. True. Since this power series is centered at a = 0 and it diverges at x = 6, its radius of
convergence must be R ≤ 6. Therefore, it must diverge at x = 10 since 10 > R.

P P
5. If an is divergent, then |an | is divergent.

Solution. True. This is the contrapositive of the fact that absolute convergence implies convergence.
Math 222 Chapter 11 Review 05/01/19. [p.2]

Determine whether each sequence is convergent or divergent. If it is convergent, find its limit.
n o∞
2+n3
6. {an }∞
n=0 = 1+2n 3
n=0

Solution. Converges. Can do by L’Hôpital, by using facts you know about rational functions, or:

2 + n3 2 + n3 n−3 1 + 2n−3 1+0 1

lim = lim · = lim = = .
n→∞ 1 + 2n3 n→∞ 1 + 2n3 n−3 n→∞ 2 + n−3 2+0 2
 
n sin n
7. {an }∞
n=0 =
n2 + 1

Solution. Converges by squeeze theorem. We have

n sin(n) n
0≤ ≤ 2 → 0,
n2 + 1 n +1
n sin n
so limn→∞ n2 +1 = 0.
( 4n )∞
3
8. {an }∞
n=1 = 1+
n
n=1

Solution. We write  4n  4n

3 12
1+ = 1+ .
n 4n
Now the base of the fraction and the exponent match, so we can use our formula limn→∞ (1 + nx )n = ex .
We have:  4n  4n
3 12
lim 1 + = lim 1 + = e12 .
n→∞ n n→∞ 4n

9. A sequence is defined recursively by a1 = 1, an+1 = 31 (an + 4). Using that {an } is increasing and
an < 5 for all n, deduce that {an } is convergent, and find its limit.

Solution. The limit exists by the monotone convergence theorem. Let L = limn→∞ an . We take the
limit of both sides of the equation an+1 = 31 (an + 4). We have

1
lim an+1 = lim (an + 4),
n→∞ n→∞ 3
so
1
L= (L + 4).
3
And we solve this nice linear equation.
1
L= (L + 4) =⇒ 3L = L + 4 =⇒ 2L = 4 =⇒ L = 2.
3
Math 222 Chapter 11 Review 05/01/19. [p.3]

Determine whether each series is convergent or divergent. If it is convergent, find its sum.
∞
X 1
10.
n=1
n(n + 3)

Solution. This is a telescoping sum. We use partial fractions. We set

1 A B
= + ,
n(n + 3) n n+3

and obtain
1 = A(n + 3) + Bn.
Setting n = 0, we have A = 1/3. Setting n = −3, we have B = −1/3. Hence

1 1/3 −1/3
= + ,
n(n + 3) n n+3

and we write out terms to see what cancels in the partial sums.
K K  
X 1 X 1/3 −1/3
= +
n=1
n(n + 3) n=1 n n+3
     
1/3 −1/3 1/3 −1/3 1/3 −1/3
= + + + + +
1 4 2 5 3 6
   
1/3 −1/3 1/3 −1/3
+ + + ... + +
4 7 K K +3
1/3 1/3 1/3 −1/3
= + + + .
1 2 3 K +3
Therefore
∞ K  
X 1 X 1 1/3 1/3 1/3 −1/3 11
= lim = lim + + + = .
n=1
n(n + 3) K→∞ n=1 n(n + 3) K→∞ 1 2 3 K +3 18

∞
X
11. (arctan(n + 1) − arctan(n))
n=1

Solution. Let us write Sn for the n-th partial sum. By telescoping cancellation, we have

Sn = (arctan(2) − arctan(1)) + (arctan(3) − arctan(2)) + . . . + (arctan(n + 1) − arctan(n))

π
= arctan(n + 1) − arctan(1) = arctan(n + 1) − .
4
Therefore
∞
X π π π π
(arctan(n + 1) − arctan(n)) = lim Sn = lim (arctan(n + 1) − )= − = .
n=1
n→∞ n→∞ 4 2 4 4
Math 222 Chapter 11 Review 05/01/19. [p.4]

e2 e3 e4
12. 1 − e + − + − ···
2! 3! 4!

Solution. This is an alternating series, and we see that we can write an explicit formula for this
series. We have
∞ ∞
e2 e3 e4 X (−1)n en X (−e)n
1−e+ − + − ··· = = .
2! 3! 4! n=0
n! n=0
n!
We recognize this as the Taylor series of ex (centered at 0) with x = −e plugged in. Therefore
∞
e2 e3 e4 X (−e)n
1−e+ − + − ··· = = e−e .
2! 3! 4! n=0
n!

Determine whether each series is conditionally convergent, absolutely convergent, or divergent.

∞
X n3
13.
n=1
5n

Solution. Converges absolutely by ratio test. We have:

an+1 (n + 1)3 /5n+1
lim = lim
n→∞ an n→∞ n3 /5n
 3
n+1 1 1
= lim · = < 1.
n→∞ n 5 5
So the ratio test says the series converges absolutely.
∞  
X n
14. ln
n=1
3n + 1

n
Solution. Diverges by n-th term test. Since limn→∞ 3n+1 = 13 , we have
   
n 1
lim ln = ln 6= 0,
n→∞ 3n + 1 3
so the sum is divergent.
∞ √
X n
15. (−1)n−1
n=1
n+1

Solution. This is conditionally convergent by A.S.T.

First, we show that the series convergent by A.S.T. To do so, we have to check that the positive
1/2 1/2
part of the series bn = nn+1 is decreasing. To do so, we show the function f (x) = xx+1 is decreasing,
by showing its derivative is negative. We have
1 √
(1/2)x−1/2 (x + 1) − x−1/2 √
2 x
− 12 x
0
f (x) = = <0
(x + 1)2 (x + 1)2
1 1√
⇔ √ − x < 0 ⇔ 1 − x < 0 ⇔ x > 1.
2 x 2
√
Moving from the first line to the second above here, we multiply by 2 x. Thus, we find that f is
decreasing when x > 1.
Math 222 Chapter 11 Review 05/01/19. [p.5]

Next, we see that

√
(−1)n−1 n n1/2 1
lim = lim = lim 1/2 = 0.
n→∞ n+1 n→∞ n + 1 n→∞ n + n−1/2

We see that this final limit is 0 because n1/2 approaches +∞ as n → ∞, and n−1/2 approaches 0 as
n → ∞. Having checked the assumptions of the A.S.T., we see that the series is convergent.

Now we show that the series is not absolutely convergent, meaning that the series
∞ √ ∞ √
X (−1)n−1 n X n
=
n=1
n + 1 n=1
n + 1
√
n √1 .
is divergent. This is divergent by L.C.T. with bn = n = n
We have

an n
lim = lim = 1,
n→∞ bn n→∞ n + 1

P∞ √
P P 1 n
and bn = n1/2
is a divergent p-series. Therefore by the L.C.T., n=1 n+1 diverges, so the original
series is not absolutely convergent.
∞
X (−1)n+1
16. How many terms of the series are needed to approximate the sum with absolute value of
n=1
n5
the error ≤ 0.00001?

Solution. Using the alternating series error, we have:

1 1
nth term error ≤ 5
≤ 0.00001 = 5
(n + 1) 10

⇔ (n + 1)5 ≥ 105 ⇔ n + 1 ≥ 10 ⇔ n ≥ 9.
So we need at least 9 terms.
Math 222 Chapter 11 Review 05/01/19. [p.6]

Find the radius of convergence and the interval of convergence of the power series.
∞
X 2n (x − 3)n
17. √
n=0
n+3

Solution. Let’s ask our friend, the ratio test, what it thinks about this power series. We have
√
an+1 2n+1 (x − 3)n+1 n+3
lim = lim √ · n
n→∞ an n→∞ n+4 2 (x − 3)n
 1/2
n+3
= lim · 2 · |x − 3| = 2|x − 3|.
n→∞ n + 4

So the ratio test tells us that the series is absolutely convergent when
1
2|x − 3| < 1 ⇔ |x − 3| < ,
2
i.e. on the interval (3 − 12 , 3 + 21 ) = (2.5, 3.5).

Now let us check the endpoints. If we plug in x = 2.5 = 3 − 12 , we obtain the series
∞ ∞ ∞
X 2n (3 − 12 − 3)n X 2n ( −1 )n X (−1)n
√ = √ 2 = √ .
n=0
n+3 n=0
n+3 n=0
n+3

This is conditionally convergent by A.S.T. and L.C.T. [Check yourself!] Similarly, if we plug in x = 3.5,
we obtain
∞
X 1
√ ,
n=0
n +3

which is divergent. So the interval of convergence is [2.5, 3.5), and the radius is 12 .
Math 222 Chapter 11 Review 05/01/19. [p.7]

Find the Taylor series the function and its radius of convergence. Then find the degree n
Taylor polynomial and estimate the error of the approximation f (x) ∼ Tn (x) for x in the given
interval.
18. Find the Taylor series of f (x) = ln(4 − x), centered at a = 0, and find its radius of convergence.
Then find T4 (x), the degree 4 Taylor polynomial, and approximate the error of the approximation
f (x) ∼ T4 (x) in the interval [−2, 2].

Solution. Power series techniques give us this function’s Taylor series. We have
∞     ∞  
0 −1 (−1/4) X −1 x n X −1
f (x) = = = · = n+1
xn .
4−x 1 − (x/4) n=0 4 4 n=0
4

This geometric series converges on the interval |r| = |x/4| < 1 ⇔ |x| < 4, so the radius of convergence
is 4. Hence
∞   ∞  
−1 −1
Z X X
n
f (x) = n+1
x dx = n+1
xn+1 + C.
n=0
4 n=0
(n + 1)4

Plugging in x = 0, we obtain ln(4) = C, so the Taylor series of f (x) is

∞  
X −1
xn+1 + ln(4).
n=0
(n + 1)4n+1

The radius of converges is the same as the radius of convergence of its antiderivative, which is 4.

T4 (x) is all terms of this Taylor series up to and including the x4 term.
1 1 1 3 1 4
T3 (x) = ln(4) − x − x2 − x − x .
4 32 192 1024

−24
Now we compute the Taylor series error. The 5-th derivative of f (x) is f (5) (x) = (4−x)5 . The
(5)
function f (x) is decreasing and negative on the interval [−2, 2], so its maximum in absolute value
on [−2, 2] occurs at the right endpoint x = +2. [The function 4 − x is decreasing. Then we take the
reciprocal, then we take a negative sign]. By the Taylor series error formula:

max[−2,2] |f (5) | · |x − 0|5 | − 24/25 | · 25 24 1

|R4 (x)| ≤ ≤ = = .
5! 5! 5! 5
The biggest the error can be is 1/5.
Math 222 Chapter 11 Review 05/01/19. [p.8]

sin x − x
19. Use series to evaluate lim .
x→0 x3

Solution. Using the Taylor series formula of sin.

x3 x5 x7
sin(x) = x − + − + ....
3! 5! 7!
So
x3 x5 x7
sin(x) − x = − + − + ...,
3! 5! 7!
and then
3 5 7
x
sin(x) − x + x5! − x7! + . . .
3
= − 3!
x x3
2
−1 x x4
= + − + ....
3! 5! 7!
Taking the limit as x → 0 of the power series, we see that we can plug in x = 0 directly. Therefore

−1 x2 x4
 
sin x − x −1 −1
lim 3
= lim + − + ... = +0= .
x→0 x x→0 3! 5! 7! 3! 6