Infinite Series Part 2

Infinite Series
(Part-2)
P. Sam Johnson
May 7, 2023
P. Sam Johnson (NIT Karnataka) Infinite Series (Part-2) May 7, 2023 1 / 156
Alternating Series
We now discuss convergence of series whose terms are alternately positive

and negative.
Definition 1.
A series which the terms are alternately positive and negative is an
alternating series.
Example 2.
1 1 1 1(−1)n+1
1. 1 − 2 + 3 − 4 + 5 − ··· +
n + ···
n
2. −2 + 1 − 21 + 14 − 18 + · · · + (−1)
2n
4
+ ···
3. 1 − 2 + 3 − 4 + 5 − 6 + · · · + (−1) n + n+1 ···
Alternating Series
We see from these examples that the nth term of an alternating series is of
the form
an = (−1)n+1 un or an = (−1)n un
where un = |an | is a positive number.
Series (1), called the alternating harmonic series, converges, as we will

see in a moment. Series (2), a geometric series with ratio r = −1/2,
converges to −2/[1 + (1/2)] = −4/3. Series (3) diverges because the nth
term does not approach zero.
We prove the convergence of the alternating harmonic series by applying

the Alternating Series Test. This test is for convergence of an alternating
series and cannot be used to conclude that such a series diverges. If we
multiply (u1 − u2 + u3 − u4 + · · · ) by −1, we see that the test is also valid
for the alternating series −u1 + u2 − u3 + u4 − · · · , as with the one in
Series (2) given above.
The Alternating Series Test
Theorem 3 (The Alternating Series Test (Leibniz’s Theorem)).
The Series
∞
X
(−1)n+1 un = u1 − u2 + u3 − u4 + · · ·
n=1
converges if all three of the following conditions are satisfied:

1. The un ’s are all positive.
2. un ≥ un+1 for all n ≤ N, for some integer N.
3. un → 0.
Proof of Leibniz’s Theorem
If n is an even integer, say n = 2m, then the sum of the first n terms is
s2m = (u1 − u2 ) + (u3 − u4 ) + · · · + (u2m−1 − u2m )

= u1 − (u2 − u3 ) − (u4 − u5 ) − · · · − (u2m−2 − u2m−1 ) − u2m .
The first equality shows that s2m is the sum of m nonnegative terms, since
each term in parentheses is positive or zero.
Hence s2m+2 ≥ s2m , and the sequence {s2m } is non-decreasing and

bounded from above, it has a limit, say
lim s2m = L.
m→∞
Proof of Leibniz’s Theorem (contd...)
If n is an odd integer, say n = 2m + 1, then the sum of the first n terms is

s2m+1 = s2m + u2m+1 . Since un → 0,
lim u2m+1 = 0
m→∞
and, as m → ∞,
s2m+1 = s2m + u2m+1 → L + 0 = L.
We have the following result for sequences :

For a sequence {an } the terms of even index are denoted by a2k and the
terms of odd index by a2k+1 . If a2k → L and a2k+1 → L, then an → L.
As s2m+1 → L and s2m → L, limn→∞ sn = L.
Eample
Example 4 (The alternating harmonic series).
∞
X 1 1 1 1
(−1)n+1 = 1 − + − + ···
n 2 3 4
n=1
satisfies the three requirements of Leibniz’s Theorem with N = 1; it

therefore converges by the Alternating Series Test. Notice that the test
gives no information about what the sum of the series might be.
The following figure shows histograms of the partial sums of the divergent
harmonic series and those of the convergent alternating harmonic series.
It turns out that the alternating harmonic series converges to ln 2. Rather
than directly verifying the definition un ≥ un+1 , a second way to show that
the sequence {un } is nonincreasing is to define a differentiable function
f (x) satisfying f (n) = un . That is, the values of f match the values of the
sequence at every positive integer n. If f 0 (x) ≤ 0 for all x greater than or
equal to some positive integer N, then f (x) is nonincreasing for x ≥ N. It
follows that f (n) ≥ f (n + 1), or un ≥ un+1 , for n ≥ N.
Graphical Interpretation
An Example of Nonincreasing Sequence
Example 5.
We show that the sequence un = 10n/(n2 + 16) is eventually
nonincreasing. Define f (x) = 10x/(x 2 + 16).
Then from the Derivative Quotient Rule,
10(16 − x 2 )
f 0 (x) = ≤ 0 whenever x ≥ 4.
(x 2 + 16)2
It follows that un ≥ un+1 for n ≥ 4. That is, the sequence {un } is

nonincreasing for n ≥ 4.
A graphical interpretation of the partial sums shows how an alternating

series converges to its limit L when the three conditions are satisfied with
N = 1. Starting from the origin of the x-axis, we lay off the positive
distance s1 = u1 .
To find the point corresponding to s2 = u1 − u2 , we back up a distance

equal to u2 . Since u2 ≤ u1 , we do not back up any farther than the origin.
We continue in this seesaw fashion, backing up or going forward as the

signs in the series demand. But for n ≥ N, each forward or backward step
is shorter than (or at most the same size as) the preceding step, because
un+1 ≤ un . And since the nth term approaches zero as n increases, the
size of step we take forward or backward gets smaller and smaller.
We oscillate across the limit L, and the amplitude of oscillation approaches

zero. The limit L lies between any two successive sums sn and sn+1 and
hence differs from sn by an amount less than un+1 .
Because |L − sn | < un+1 for n ≥ N,
We can make useful estimates of the sums of convergent alternating series.
The Alternating Series Estimation Theorem
Theorem 6 (The Alternating Series Estimation Theorem).
If the alternating series
∞
X
(−1)n+1 un
n=1
satisfies the three conditions of Leibniz’s Theorem, then for n ≥ N,
sn = u1 − u2 + · · · + (−1)n+1 un .
approximates the sum L of the series with an error whose absolute value is
less than un+1 , the numerical value of the first unused term. Furthermore,
the remainder, L − sn , has the same sign as the first as the first unused
term.
Example
Example 7.
Consider the following series
∞
X 1 1 1 1 1 1 1 1 1
(−1)n n
=1− + − + − + − + − ··· .
2 2 4 8 16 32 64 128 256
n=0
The Alternating Series Estimation Theorem says that if we truncate the

series after the eighth term, we throw away a total that is positive and less
than 1/256. The sum of the first eight terms is 0.6640625. The sum of
the series is
1 1 2
= = .
1 − (−1/2) 3/2 3
The difference, (2/3) − 0.6640625 = 0.0026041666 · · · is positive and less
than (1/256) = 0.00390625.
The alternating harmonic series does not converge absolutely. The
corresponding series of absolute value is the (divergent) harmonic series.
Conditionally Convergent
If we replace all the negative terms in the alternating series in
∞
X 1
(−1)n ,
n=0
2n
1/2n . The original

P
changing them to positive terms instead, we obtain the geometric series
series and the new series of absolute values both converge (although to different sums).
For an absolutely convergent series, changing infinitely many of the negative terms in the series
to positive values does not change its property of still being a convergent series. Other
convergent series may behave differently. The convergent alternating harmonic series has
infinitely many negative terms, but if we change its negative terms to positive values, the
resulting series is the divergent harmonic series. So the presence of infinitely many negative
terms is essential to the convergence of the alternating harmonic series. The following
terminology distinguishes these two types of convergent series.
Conditionally Convergent
Definition 8 (Conditionally Convergent).
A series that converges but does not converge absolutely is called a
conditionally convergent series.
The alternating harmonic series is conditionally convergent, or converges

conditionally. The next example extends that result to the alternating
p-series.
Alternating p-Series
Example 9.
If p is a positive constant, the sequence {1/np } is a decreasing sequence
with limit zero. Therefore the alternating p-series
∞
X (−1)n−1 1 1 1
=1− + p − p + ··· , p>0
np 2p 3 4
n=1
converges.
If p > 1, the series converges absolutely. If 0 < p ≤ 1, the series converges

conditionally.
Conditional convergence: 1 − √1 + √1 − √1 + ···.

2 3 4
1 1 1
Absolute convergence: 1 − 23/2
+ 33/2
− 43/2
+ ....
Summary of Facts about Alternating Series
P
If an is an alternating series, then exactly one of the following holds:
P
an is convergent P(can be concluded
P by absolute convergence test,
in which case both |an | and an are convergent).
P P
P an is conditionally convergent (i.e., an is convergent, while
|an | is not).
P P P
an is divergent (in which case, both an and |an | are
divergent. nth -term Test may be helpful most of the times.)
Rearranging Series
We need to be careful when using a conditionally convergent series. We

have seen with the alternating harmonic series that altering the signs of
infinitely many terms of a conditionally convergent series can change its
convergence status. Even more, simply changing the order of occurrence
of infinitely many of its terms can also have a significant effect, as we now
discuss.
We can always rearrange the terms of a finite collection of numbers

without changing their sum. The same result is true for an infinite series
that is absolutely convergent.
Rearranging Series
Absolutely convergence is important for two reasons. First, we have good

tests for convergence of series of positive terms. Second, if a series
converges absolutely, then it converges. That is the thrust of the next
theorem.
Theorem 10 (The Rearrangement Theorem for Absolutely
Convergent Series).
If ∞
P
n=1 an converges absolutely,
P∞ and b1 , b2 , . . . , bn , . . . is any arrangement
of the sequence {an }, then n=1 bn converges absolutely and
∞
X ∞
X
bn = an .
n=1 n=1
Rearranging Series
On the other hand, if we rearrange the terms of a conditionally convergent

series, we can get different results. In fact, for any real number r , a given
conditionally convergent series can be rearranged so its sum is equal to r .
Here is an example of summing the terms of a conditionally convergent

series with different orderings, with each ordering giving a different value
for the sum.
Rearranging Series
Example 11.
We know that the alternating harmonic series ∞ n+1 /n converges to some number L.
P
n=1 (−1)
Moreover, by Theorem 10, L lies between the successive partial sums s2 = 1/2 and s3 = 5/6, so
L 6= 0. If we multiply the series by 2 we obtain
∞
(−1)n+1

X 1 1 1 1 1 1 1 1 1 1
2L = 2 =2 1− + − + − + − + − + − ···
n=1
n 2 3 4 5 6 7 8 9 10 11
2 1 2 1 2 1 2 1 2
=2−1+ − + − + − + − + − ··· .
3 2 5 3 7 4 9 5 11
Now we change the order of this last sum by grouping each pair of terms with the same odd
denominator, but leaving the negative terms with the even denominators as they are placed (so
the denominators are the positive integers in their natural order). This rearrangement gives

1 2 1 1 2 1 1 2 1 1
(2 − 1) − + − − + − − + − − + ···
2 3 3 4 5 5 6 7 7 8

1 1 1 1 1 1 1 1 1 1
= 1− + − + − + − + − + − ···
2 3 4 5 6 7 8 9 10 11
∞
X (−1)n+1
= = L.
n=1
n
Rearranging Series
So by rearranging the terms of the conditionally convergent series

∞
X 2(−1)n+1
,
n
n=1
the series becomes

∞
X (−1)n+1
,
n
n=1
which is the alternating harmonic series itself. If the two series are the
same, it would imply that 2L = L, which is clearly false since L 6= 0.
What is wrong here?
Exercise 12.
Multiplying both sides of the alternating harmonic series
1 1 1 1 1 1 1 1 1 1 1
S =1− + − + − + − + − + − + ...
2 3 4 5 6 7 8 9 10 2 = 11 12
by 2 to get
2 1 2 1 1 2 1 2 1 2 1
2S = 2 − 1 + − + − + + − + − + − + ....
3 2 5 3 3 7 4 9 5 11 6
Collect terms with the same denominator, as the arrows indicate, to arrive at
1 1 1 1 1
2S = 1 − + − + − + ....
2 3 4 5 6
The series on the right-hand side of this equation is the series we started with.
Therefore,2S = S. and dividing by S gives 2 = 1. (Source: “Riemann’s Rearrangement Theorem
“by Stewart Galanor, Mathematics Teacher,Vol.80,No.8,1987, pp. 675-681.)
Rearranging Series
Example 11 shows that we cannot rearrange the terms of a conditionally

convergent series and expect the new series to be the same as the original
one.
When we use a conditionally convergent series, the terms must be added

together in the order they are given to obtain a correct result.
In contrast, Theorem 10 guarantees that the terms of an absolutely

convergent series can be summed in any order without affecting the result.
Outline of the proof of the The Rearrangement Theorem
for Absolutely Convergent Series
Part (a)
Let ε be a positive real number, let L = ∞

P Pk
n=1 an , and let sk = n=1 an .
Show that for some index N1 and for some index N2 ≥ N1 ,
P∞ ε ε
n=N1 |an | < 2 and |sN2 − L| < 2 .
Pnin the sequence {bn },

Since all the terms a1 , a2 , . . . , aN2 appear somewhere
there is index N3 ≥ N2 such that if n ≥ N3 , then ( k=1 bk ) − sN2 is at
most a sum of terms am with m ≥ N1 . Therefore if n ≥ N3 ,
| nk=1 bk − L| ≤ | nn=1 bk − sN2 | + |sN2 − L|

P P
≤ ∞
P
k=N1 |ak | + |sN2 − L| < ε.
Outline of the proof of the The Rearrangement Theorem
for Absolutely Convergent Series (contd...)
Part (b)
The argument in part (a)Pshows thatPif ∞

P
n=1 an converges absolutely then
P ∞ ∞ ∞
b
n=1 n converges and b
n=1 n = n=1 an .
Now show that because ∞

P P∞
n=1 an converges, n=1 bn converges to
P ∞
a
n=1 n .
Applying the Rearrangement Thoerem
Example 13 (Applying the Rearrangement Thoerem).
The series
1 1 1 1
+ −
1− + · · · + (−1)n−1 2 + · · ·
4 9 16 n
converges absolutely.
A possible rearrangement of the terms of the series might start with a

positive term, then two negative terms, then three positive terms, then
four negative terms, and so on: After k terms of one sign, take k + 1 terms
of the opposite sign. The first ten terms of such a series look like this:
1 1 1 1 1 1 1 1 1
1− − + + + − − − − + ··· .
4 16 9 25 49 36 64 100 144
The Rearrangement Theorem says that both series converge to the same
value.
In this example, if we had the second series to begin with, we would

probably be glad to exchange it for the first, if we knew that we could.
We can do even better: The sum of either series is also equal to

∞ ∞
X 1 X 1
− .
(2n − 1)2 (2n)2
n=1 n=1
If we rearrange infinitely many terms of conditionally convergent series, we

can get result that are far different from the sum of the original series.
Here is an example.
Example 14 (Rearranging the Alternating Harmonic Seires).

The alternating harmonic seires
1 1 1 1 1 1 1 1 1 1 1
− + − + − + − + − + − ···
1 2 3 4 5 6 7 8 9 10 11
can be rearranged to diverge to reach any preassigned sum.
Rearranging the Alternating Harmonic Seires
∞
X
(−1)n+1 /n to diverge. The series of terms
P 1
Rearranging 2n−1
n=1 P −1
diverges to +∞ and the series of terms 2n diverges to −∞. No matter
how far out in the sequence of odd-numbered terms we begin, we can
always add enough positive terms to get an arbitrarily large sum. Similarly,
with the negative terms, no matter how far out we start, we can add
enough consecutive even-numbered terms to get a negative sum of
arbitrarily large absolute value. If we wished to do so, we could start
adding odd-numbered terms until we had a sum greater than +3, say, and
then follow that with enough consecutive negative terms to make the total
less than -4. We could then add enough positive terms to make the total
greater than +5 and follow with consecutive unused negative terms to
make a new total less than −6, and so on. In this way, we could make the
swings arbitrarily large in either direction.
∞
X
Rearranging (−1)n+1 /n to converge to 1. Another possibility is to
n=1
focus on a particular limit. Suppose we try to get sums that converge to 1.
We start with the first term, 1/1, and then subtract 1/2. Next we add 1/3
and 1/5, which brings the total back to 1 or above. Then we add
consecutive negative terms until the total is less than 1. We continue in
this manner: When the sum is less than 1, add positive terms until the
total is 1 or more: then subtract (add negative) terms until the total is
again less than 1.
This process can be continued indefinitely. Because both the

odd-numbered terms and the even-numbered terms of the original series
approach zero as n → ∞.
The amount by which our partial sums exceed 1 or fall below it approaches
zero. So the new series converges to 1. The rearranged series starts like
this:
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
− + + − + + − + + − + + − + + − + + − + − +···
1 2 3 5 4 7 9 6 11 13 8 15 17 10 19 21 12 23 25 14 27 16
The kind of behavior illustrated by the series the above example is typical
of what happen with any conditionally convergent series. Therefore we
must always add the terms of a conditionally convergent series in the order
given.
Summary
We have now developed several tests for convergence and divergence of

series. In summary:
1. The nth-Term Test: Unless an → 0, the series diverges.

P n
2. Geometric seies: ar converges if |r | < 1; otherwise it diverges.
p
P
3. p-series: 1/n converges if p > 1; otherwise it diverges.
4. Series with nonnegative terms: Try the Integral Test, Ratio Test,
or Root Test. Try comparing to a known series with the Comparison
Test.
P
5. SeriesPwith some negative terms: Does |an | converge? If yes, so
does an , since absolute convergence implies convergence.
P
6. Alternating Series: an converges if the series satisfies the
conditions of the Alternating Series Test.
Exercise
Exercise 15 (Determining Convergence or Divergence).
Which of the following alternating series converge, and which diverge?
Give reasons for your answers.
∞ ∞
X
n+1 1 X
1. (−1) 6. (−1)n+1 (0.1)n
n=1
n3/2 n=1
∞ h i
P (−1)n+1 sin nx ∞
2. n3 X (−1)n
n=1 7. √
∞ 1+ n
(−1)n csch n
P
3. n=1
n=1 ∞
∞ X n!
X
n 1 8. (−1)n+1
4. (−1) ln 1 + 2n
n n=1
n=1
∞ √
X 3 n+1 9. 1 + 14 − 19 − 16 1
+ 1
+ 1
−
5. (−1)n+1 √ 1 1
25 36
n=1
n+1 49 − 64 + · · ·
Determining Convergence or Divergence
Exercise 16.
∞ ∞
X √ X (−1)n+1 (n!)2
(−1)n+1 ( 10)
n
1. 7.
(2n)!
n=1 n=1
∞
X 1 ∞
2. (−1)n+1 X (2n)!
n ln n 8. (−1)n
n=2
∞ 2n !n
n=1
X n
3. (−1)n ∞
n+1 X p
n=1
∞ 9. (−1)n ( n2 + n − n)
X
4. (−5)−n n=1
n=1 ∞
√ √
q
∞
X
n
X (−1)n−1 10. (−1) ( n + n − n)
5.
n2 + 2n + 1 n=1
n=1
∞ ∞
X cos nπ X
6. 11. (−1)n sech n
n
n=1 n=1
Exercises
Exercise 17.
Determine how many terms should be used to estimate the sum of the
entire series with an error of less than 0.001.
∞
(−1)n n21+3
P
1.
n=1
∞
1
(−1)n ln(ln(n+2))
P
2.
n=1
Exercises
Exercise 18.
Use any method to determine whether the series converges or diverges.
Give reasons for your answer.
∞ ∞
1 1 ln n 3
P P
1. 2n+1 − 2n+2 7. n
n=1 n=2
∞ ∞
P 3
2. 10+n4/3 8.
P 1
n=2 1+2+22 +···+2n
∞ n=2
2
2 n ∞
P
3. 1− n
P 1+3+32 +···+3n−1
n=1 9. 1+2+3+···+n
∞ n n=2
n−2
− 32
P
4. n2 +3n ∞
en
(−1)n
P
n=1 10.
1 1 e n +e n2
5. 2 − 2 + 12 − 12 + 1
2 − 1
2 + ··· n=0
∞ ∞
4·6·8···(2n)
sin √1n
P P
6. 11. 5n+1 (n+2)!
n=3 n=1
Error Estimation
Exercise 19.
Estimate the magnitude of the error involved in using the sum of the first
four terms to approximate the sum of the entire series.
∞
X 1
1. (−1)n+1
n
n=1
∞
X (0.01)n
2. (−1)n+1
n
n=1
∞
X
3. 1
1+t = (−1)n t n , 0 < t < 1
n=1
Exercise
Exercise 20.
Approximate the sums in
∞
X 1
(−1)n
(2n)!
n=1
with an error of magnitude less than 5 × 10−6 .
Exercise
Exercise 21.
(a) The series 13 − 12 + 19 − 14 + 27
1
− 18 + . . . + 31n − 21n + . . .
Does not meet one of the conditions of Leibniz’s Theorem. Which
one?
(b) Find the sum of the series in part (a).
Exercises
Exercise 22.
The limit L of an alternating series that satisfies the conditions of
Theorem 3 lies between the values of any two consecutive partial sums.
This suggests using the average
sn + sn+1 1
= sn + (−1)n+2 an+1
2 2
to estimate L. Compute
1 1
· s20 +
2 21
as an approximation to the sum of the alternating harmonic series. The
exact sum is ln 2 = 0.69314718 . . ..
Exercise
Exercise 23 (The sign of an alternating series that satisfies
the conditions of the Leibniz’s Theorem).
Prove the assertion in the Alternating Series Estimation Theorem that
whenever an alternating series satisfying the conditions of Leibniz’s
Theorem is approximated with one of its partial sums, then the remainder
(sum of then unsaved terms) has the same sign as the first unused term.
(Hint: Group the remainder’s terms in consecutive pairs.)
Exercise
Exercise 24.
Show that the sum of the first 2n terms of the series
1 1 1 1 1 1 1 1 1
1− + − + − + − + − + ...
2 2 3 3 4 4 5 5 6
Is the same as the sum of the first n terms of the series
1 1 1 1 1
+ + + + + ....
1.2 2.3 3.4 4.5 5.6
Do these series converge? What is the sum of the first 2n + 1 terms of the
first series? If the series converge, what is their sum?
Exercise
Exercise 25.
Show that
∞
X
an
n=1
diverges, then
∞
X
|an |
n=1
diverges.
Exercise
Exercise 26.
Show that if
∞
X
an
n=1
converges absolutely, then

X∞ X ∞
an ≤ |an |.

n=1 n=1
Exercise
Exercise 27.
∞
X ∞
X
Show that if an and bn both converge absolutely, then so does
n=1 n=1
∞
X
(a) (an + bn )
n=1
∞
X
(b) (an − bn )
n=1
∞
X
(c) kan (k any number)
n=1
Exercise
Exercise 28.
Show by example that
∞
X
an b n
n=1
∞
X ∞
X
may diverge even if an and bn both converge.
n=1 n=1
Exercises
Exercise 29.
an2 converges.
P P
1. If an converges absolutely, prove that
2. Does the series
∞
X 1 1
−
n=1
n n2
converge or diverge? Justify your answer.

3. In the alternating harmonic series, suppose the goal is to arrange the terms to get a new
series that converges to −1/2. Start the new arrangement with the first negative term,
which is −1/2. Whenever you have a sum that is less than or equal to −1/2, start
introducing positive terms, taken in order, until the new total is greater than −1/2. Then
add negative terms until the total is less than or equal to −1/2 again. Continue this
process until your partial sums have been above the target at least three times and finish
at or below it. If sn is the sum of the first n terms of your new series, plot the points
(n, sn ) to illustrate how the sums are behaving.
Power Series
Power Series
We have discussed so far some tests for convergence of infinite series. Now
we are going to see a special series and see that its sum looks like “infinite
polynomials.”
We call these sums “power series” because they are defined as infinite
series of powers of some variable, in our case x.
Like polynomials, power series can be added, subtracted, multiplied,

differentiated, and integrated to give new power series.
Power Series, Center, Coefficients
A power series about x = 0 is a series of the form

∞
X
cn x n = c0 + c1 x + c2 x 2 + · · · + cn x n + · · · . (1)
n=0
A power series about x = a is a series of the form

∞
X
cn (x − a)n = c0 + c1 (x − a) + c2 (x − a)2 + · · · + cn (x − a)n + · · · (2)
n=0
in which the center a and the coefficients c0 , c1 , c2 , . . . , cn , . . . are

constants.
Equation (7) is the special case obtained by taking a = 0 in Equation (2).
Power Series and Convergence
Example 30 (Geometric Series).
Taking all the coefficients to be 1 in Equation (7) gives the geometric
power series
X∞
xn = 1 + x + x2 + · · · + xn + . . . .
n=0
This is the geometric series with first term 1 and ratio x.
It converges to 1/(1 − x) for |x| < 1. We express this fact by writing
1
= 1 + x + x2 + · · · + xn + · · · , −1 < x < 1. (3)
1−x
Up to now, we have use Equation (3) as a formula for the sum of the
series on the right.
We think of the partial sums of the series on the right as polynomials

Pn (x) that approximate the function on the left.
For values of x near zero, we need take only a few terms of the series to
get a good approximation. As we move toward a = 1, or −1, we must
take more terms.
The following figure show the graphs of f (x) = 1/(1 − x), and the
approximating polynomials yn = Pn (x) for n = 0, 1, 2 and 8. The function
f (x) = 1/(1/x) is not continuous on intervals containing x = 1. where it
has a vertical asymptote. The approximations do not apply when x ≥ 1.
Example 31 (A Geometric Series).
The power series
n
1 1 1 2 1
− (x − 2) + (x − 2) + · · · + − (x − 2)n + . . . (4)
4 2 4 2
matches equation (2) with
a = 2, c0 = 1, c1 = −1/2, c2 = 1/4, · · · , cn = (−1/2)n .
This is a geometric series with first term 1 and ratio r = − x−2

2 .
The series converges for | x−2

2 | < 1 or 0 < x < 4.
1 1 2
The sum is 1−r = 1+ x−2
= x so
2
n
(x − 2)2

2 (x − 2) 1
=1− + − ··· + − (x − 2)n + . . . , 0 < x < 4.
x 2 4 2
Series (4) generates useful polynomial approximations of f (x) = 2/x for

values of x near 2:
P0 (x) = 1
1 x
P1 (x) = 1 − (x − 2) = 2 −
2 x
1 1 3x x2
P2 (x) = 1 − (x − 2) + (x − 2)2 = 3 − +
2 4 2 4
and so on.
Example 32 (Testing for Convergence Using the Ratio Test).
For what values of x does the following power series converge?
∞
X xn x2 x3
(−1)n−1 =x− + − ···
n 2 3
n=1
P
Apply the Ratio Test to the series |un |, where un is the nth term of the
un +1 n
series in question : un = n+1 |x| → |x|.
The series converges absolutely for |x| < 1. It diverges if |x| > 1 because
the nth term does not converge to zero. At x = 1, we get the alternating
harmonic series 1 − 1/2 + 1/3 − 1/4 + · · · , which converges. At x = −1
we get −1 − 1/2 − 1/3 − 1/4 − · · · , the negative of the harmonic series:
it diverges.
The given series converges for −1 < x ≤ 1 and diverges elsewhere.

∞
X x 2n−1 x3 x5
(−1)n−1 =x− + − ···
2n − 1 3 5
n=1
P
Apply the Ratio Test to the series |un |, where un is the nth term of the
un +1 2n−1 2
series in question : un = 2n+1 x → x 2 . The series converges absolutely
for x 2 < 1. It diverges for x 2 < 1 becuase the nth term does not converge
to zero. At x = 1 the series becomes 1 − 1/3 + 1/5 − 1/7 + · · · , which
converges by the Alternating Series Theorem. It also converges at x = −1
because it is again an alternating series that satisfies the conditions for
convergence. The value at x = −1 is the negative of the value at x = 1.
Thus the given series converges for −1 ≤ x ≤ 1 and diverges elsewhere.
∞
X x2 x3
1+x + + + ···
2! 3!
n=1
P
Apply the Ratio Test to the n+1 |un|, where un is the nth term of the
series
|x|
series in question : unu+1
x
= (n+1)! · xn!n = n+1 → 0 for every x.

n
The series converges absolutely for all x.
∞
X
n!x n = 1 + x + 2!x 2 + 3!x 3 + · · ·
n=1
P
Apply the Ratio Test to the series n+1|un |, where un is the nth term of the
un +1 (n+1)!x
series in question : un = | n!x n = (n + 1)|x| → ∞ unless x = 0.
The Convergence Theorem for Power Series
Theorem 36 (The Convergence Theorem for Power Series).
If the power series
∞
X
an x n = a0 + a1 x + a2 x 2 + · · ·
n=0
converges for x = c 6= 0, then it converges absolutely for all x with

|x| < |c|.
If the series diverges for x = d. then it diverges for all x with |x| > |d|.
Proof of the Convergence Theorem for Power Series
Suppose the series ∞ n n

P
n=0 an c converges. Then limn→∞ an c = 0. Hence,
n
there is an integer N such that |an c | < 1 for all n ≥ N. That is,
1
|an | < for n ≥ N. (5)
|c|n
Now take any x such that |x| < |c| and consider
|a0 | + |a1 x| + · · · + |aN−1 x N−1 | + |aN x N | + |aN+1 x N+1 | + · · ·
There are only a finite number of terms prior to |aN x N |, and their sum is
finite. Starting with |aN x N |, and beyond, the terms are less than
x N x N+1 x N+2
+ + + ··· (6)

c c c
because of Inequality (5).
Proof of the Convergence Theorem for Power Series
(contd...)
But series (6) is a geometric series with ratio r = |x/c|, which is less than
1, since |x| < |c|. Hence series (6) converges, so the original series
converges absolutely. This proves the first half of the theorem.
The second half of the theorem follows from the first. If the diverges at
x = d and converges at a value x0 with |x0 | > |d|, we may tacke c = x0 in
the first half of the theorem and conclude that the series converges
absolutely at d. But the series cannot converge absolutely and diverge at
one and the same time. Hence, if it diverges at d, it diverges for all x with
|x| > |d|.
The Radius of Convergence of a Power Series
The theorem we have just proved andP the examples we have studied lead
to the conclusion that a power series cn (x − a)n behaves in one of three
possible ways.
It might converge only at x = a, or converge everywhere, or converge on

some interval of radius R centered at x = a. We prove this as a Corollary
to Convergence Theorem for Power Series.
Corollary to Convergence Theorem for Power Series
cn (x − a)n is described by one of the

P
The convergence of the series
following three possibilities:
1. There is a positive number R such that the series diverges for x with
|x − a| > R but converges absolutely for x with |x − a| < R. The
series may or may not converge at either of the endpoints x = a − R
and x = a + R.
2. The series converges absolutely for every x (R = ∞).
3. The series converges at x = a and diverges elsewhere (R = 0).
Proof of Corollary
We assume first that a = 0, so that the power series is centered at 0. If

the series converges everywhere we are in Case 2. If it converges only at
xP= 0 we are in Case 3. Otherwise there is a nonzero number dPsuch that
cn d n diverges. The set S of value of x for which the series cn x n
converges is nonempty because it contains 0 and a positive number p as
well. By Convergence Theorem for Power Series, the sereis diverges for all
x with |x| > |d|, so |x| ≤ |d| for all x ∈ S, and S is a bounded set. By the
Completeness Property of the real numbers, a nonempty, bounded set has
a least upper bound R. (the least upper bound is the smallest number with
the property that thePelements x ∈ S satisty x ≤ R.) If |x| > R ≥ p, then
x 6= S so the series n
cn x diverges.
Proof of Corollary (contd...)
If |x| < R, then |x| is not an upper bound for S (because it’s smaller than
the least upper bound) soP there is a number b ∈ S such that b > |x|.
Since b ∈ X , the series n
cn b converges and therefore the series
cn |x|n converges by Convergence Theorem for Power Series. This
P
proves the Corollary for power series centered at a = 0.
For a power series centered at a 6= 0, we set x 0 = (x − a) and repeat the

argument with 0 xP0 . Since x 0 = 0 when x = a, a radius R interval of
convergence for cn (x 0 )n centered at x 0 = 0 is the same s a radius R

cn (x − a)n centered at x = a. This
P
interval of convergence for
establishes the Corollary for the general case.
Radius of Convergence
R is called radius of convergence of the power series and the interval of

radius R centered at x = a is called the interval of convergence.
The interval of convergence may be open, closed, or half-open, depending

on the particular series. At points x with |x − a| < R, the series converges
absolutely. If the series converges for all values of x, we say its radius of
convergence is infinite. If it converges only at x = a, we say its radius of
convergence is zero.
How to Test a Power Series for Convergence
1. Use the Ratio Test (or nth-Root Test) to find the interval where the
series converges absolutely. Ordinarily, this is an open interval
|x − a| < R or a − R < x < a + R.
2. If the interval of absolute convergence is finite, test for convergence
or divergence at each endpoint. Use a Comparison Test, the Integral
Test, or the Alternating Series Test.
3. If the Interval of absolute convergence is a a − R < x < a + R, the
series diverges for |x − a| > R (it does not even converge
conditionally), because the nth term does not approach zero for those
values of x.
Term-by-Term Differentiation
A Theorem from advanced calculus says that a power series can be

differentiated term by term at each interior point of its interval of
convergence.
Theorem 37 (The Term-by-Term Differentiation Theorem).

cn (x − a)n converges forP
P
If a a − R < x < a + R for some R > 0, it
defines a function f : f (x) = ∞ n
n=0 cn (x − a) , a − R < x < a + R. Such a
function f has derivatives of all orders inside the interval of convergence.
We can obtain the derivatives by differentiating the original series term by
term:
∞
X ∞
X
0 n−1 00
f (x) = ncn (x − a) f (x) = n(n − 1)cn (x − a)n−2 ,
n=0 n=2
and so on. Each of these derived series converges at every interior point of
the interval of convergence of the original series.
Example 38 (Applying Term-by-Term Differentiation).
Find series for f 0 (x) and f 00 (x)if
∞
1 X
f (x) = = 1 + x + x2 + x3 + x4 + · · · + xn + · · · = x n , −1 < x < 1
1−x
n=0
∞
1 X
f 0 (x) = = 1 + 2x + 3x 2 + 4x 3 + · · · + nx n−1 + · · · = nx n−1 ,
(1 − x 2 ) n=1
for −1 < x < 1
∞
2 X
f ”(x) = 3
= 2 + 6x + 12x 2 + · · · + n(n − 1)x n−2 + · · · = n(n − 1)x n−2 ,
(1 − x ) n=2
for −1 < x < 1.
CAUTION
Term-by-Term differentiation might not work for other kinds of series. For
example, the trigonometric series
∞
X sin(n!x)
n2
n=1
converges for all x. But if we differentiate term by term we get the series
∞
X n!cos(n!x)
n2
n=1
which diverges for all x. This is not a power series, since it is not a sum of
positive integer powers of x.
Term-by-Term Integration
Another advanced calculate theorem states that a power series can be

integrated term by term throughout its interval of convergence.
Theorem 39 (The Term-by-Term Integration Theorem).

Suppose that f (x) = ∞ n
P
n=0 cn (x − a) converges for
a − R < x < a + R (R > 0). Then
∞
X (x − a)n+1
cn
n+1
n=1
converges for a − R < x < a + R and

∞
(x − a)n+1
Z X
f (x)dx = cn +c
n+1
n=0
for a − R < x < a + R.

Example 40 (A series for tan−1 x, −1 ≤ x ≤ 1).
Identify the function
x3 x5
f (x) = x − + − ··· , −1 ≤ x ≤ 1.
3 5
We differentiate the original series term by term and get
f 0 (x) = 1 − x 2 + x 4 − x 6 + · · · , −1 < x < 1.
This is a geometric series with first term 1 and ratio −x 2 , so

1 1
f 0 (x) = 2
= .
1 − (−x ) 1 + x2
we can now integrate f 0 (x) = 1/(1 + x 2 ) to get

Z Z
dx
f 0 (x)dx = = tan−1 x + C
1 + x2
The series for f (x) is zero when x = 0, so C = 0. Hence
x3 x5 x7
f (x) = x − + − + · · · = tan−1 x, −1 < x < 1.
3 5 7
We will see that the series also converges to tan−1 x at x = ±1.
3 5
Notice that the original series x − x3 + x5 − · · · , −1 ≤ x ≤ 1 converges
at both endpoints of the original interval of convergence, but the
“Term-by-Term Integration Theorem” can guarantee the convergence of
the differentiated series only inside the interval.
Example 41 (A series for ln(1 + x), −1 < x ≤ 1).
The series
1
= 1 + t + t2 − t3 + · · ·
1+t
converges on the open interval −1 < t < 1. Therefore,
Z x
1 t2 t3 t4 x
ln(1 + x) = dt = t − + − + ··· 0
0 1+t 2 3 4
x2 x3 x4
=x− + − + · · · , −1 < x < 1.
2 3 4
It can also be shown that the series converges at x = 1 to the number in
2, but that was not guaranteed by the theorem.
Multiplication of Power Series
Another theorem from advanced calculus state that absolutely converging

power series can be multiplied the way we multiply polynomials. We omit
the proof.
Theorem 42 (The series Multiplication Theorem for Power
Series).
If A(x) = ∞
P n
P∞ n
n=0 an x and B(x) = n=0 bn x converge absolutely for
|x| < R. and
∞
X
cn = a0 bn + a1 bn−1 + a2 bb−2 + · · · + an−1 b1 + an b0 = ak bn−k ,
k=0
P∞ n
then n=0 cn x converges absolutely to A(x)B(x) for |x| < R :
∞ ∞ ∞
! !
X X X
n n
an x · bn x = cn x n .
n=0 n=0 n=0
Proof
Suppose the series ∞ n n

P
n=0 an c converges. Then limn→∞ an c = 0. Hence,
n
there is an integer N such that |an c | < 1 for all n ≥ N. That is,
1
|an | < |c|n for n ≥ N. (5)
Now take any x such that |x| < |c| and consider
|a0 | + |a1 x| + · · · + |aN−1 x N−1 | + |aN s N | + |aN+1 x N+1 | + · · ·
There are only a finite number of terms prior to |aN x N |, and their sum is
finite. Starting with |aN x N |, and beyond, the terms are less than
x x x
| |N + | |N+1 + | |N+2 + · · · (6)
c c c
because of Inequality (5).
Multiply the Geometric series
Example 43 (Multiply the Geometric series).
P∞ n 1
n=0 x = 1 + x + x2 + · · · + xn + · · · = 1−x , for |x| < 1, by itself to get
a power series for 1/(1 − x)2 , for |x| < 1.
Let
∞
X
A(x) = an x n = 1 + x + x 2 + · · · + x n + · · · = 1/(1 − x)
n=0
∞
X
B(x) = bn x n = 1 + x + x 2 + · · · + x n + · · · = 1/(1 − x)
n=0
and
cn = a0 bn + a1 bn−1 + · · · + ak bn−k + · · · + an b0
= 1 + 1 + · · · + 1(n + 1 times) = n + 1
Then, by the Series Multiplication Theorem,

∞
X
A(x) · B(x) = cn x n
n=0
∞
X
= (n + 1)x n
n=0
= 1 + 2x + 3x 2 + 4x 3 + · · · + (n + 1)x n + · · ·
is the series for 1/(1 − x)2 .
The series all converge absolutely for |x| < 1.

d 1 1
dx 1−x = (1−x)2 .
Exercise
Exercise 44 (Intervals of Convergence).
In the following exercies, (a) find the series’ radius and interval of
convergence. For what values of x does the series converge (b) absolutely
(c) conditoinally?
P∞
1. (x + 5)n
Pn=0
∞ nx n
2. n=0 n+2
P∞ (−1)n (x+2)n
3. n=1 n
P∞ (2x+3)2n+1
4. n=0 n!
P∞ n(x+3)n
5. n=0 5n
Exercise
(c) conditoinally?
P∞ nx n
1. n=0 4n (n2 +1)
P∞ 1 n n

2. n=1 1 + n x
P∞ n
3. n=1 (ln n)x
P∞ (−1)n+1 (x+2)n
4. n=1 n2n
P∞ n (n + 1)(x − 1)n
5. (−2)
Pn=0
∞ xn
6. n=2 n(ln n)2
Exercise
(c) conditoinally?
P∞ (4x−5)2n+1
1. n=1 n3/2
P∞ (x−√2)2n+1
2. n=0 2n In the following exercises, find the series’ interval of
convergence and, within this interval, the sum of the series as a
function of x.
P∞ (x+1)2n
3. n=0 9n
P∞ √x n
4. n=0 2 − 1
P∞ x 2 −1 n
5. n=0 2
Exercise
Exercise 47.
For what values of x does the series
n
1 1 2 1
1 − (x − 3) + (x − 3) + · · · + − (x − 3)n + · · ·
2 4 2
converge? What is its sum? What series do you get if you differentiate the
given series term by term? For what values of x does the new series
converge? What it its sum?
Exercise
Exercise 48.
If you integrate the series
n
1 1 2 1
1 − (x − 3) + (x − 3) + · · · + − (x − 3)n + · · ·
2 4 2
term by term, what new series do you get? For what values of x does the
new series converge, and what is another name for its sum?
Exercise
Exercise 49.
The series
x 3 x 5 x 7 x 9 x 11
sinx = x −+ − + − + ···
3! 5! 7! 9! 11!
converges to sin x for all x.
(a) Find the first six terms of a seies for cos x. For what values of x
should the series convege?
(b) By replacing x by 2x in the series for sin x, find a series that
converges to sin 2x for all x.
(c) Using the result in part (a) and series multiplication, calculate the
first six terms of a series for 2 sin x cos x. Compare your answer with
the answer in part (b).
Exercise
Exercise 50.
The series
x2 x3 x4 x5
ex = 1 + x + + + + + ···
2! 3! 4! 5!
converges to e x for all x.
(a) Find a series for (d/dx)e x . Do you get the series for e x ?. Explain
your answer.
(b) Find a series for e x dx. Do you get the series for e x ? Explain your
R
answer.
(c) Replace x by −x in the series for e x to find a series that converges to
e x for all x. Then multyiply the series for e x and e −x to find the six
terms of a series for e −x · e x .
Exercise
Exercise 51.
The series
x 3 2x 5 17x 7 62x 9
tan x = x +
+ + + + ···
3 15 315 2835
converges to tan x for −π/2 < x < π/2.
(a) Find the first five terms of the series for ln | sec x|. For what values of
x should the series converge?
(b) Find the first five terms of the series for sec3 . Fro what values of x
should thies series converge?
(c) Check your result in part (b) by squaring the series given for sec x.
Exercise
Exercise 52.
The series
x2 5 61 6 277 8
sec x = 1 + + x4 + x + x + ···
2 24 720 8064
converges to see x for −π/2 < x < π/2.
(a) Find the first five terms of a power series for the function in
| sec x + tan x|. For what values of x should the series converge?
(b) Find the first four terms of a series for sec s tan x. For what values of
x should the series converge?
(c) Check your result in part (b) by multiplying the series for sec x by the
series given for tan x.
Exercise
Exercise 53 (Uniqueness of convergent power series).
1. Show that if two power series ∞
P n
P∞ n
n=0 an x and n=0 bn x are
convergent and equal for all values of x in an open interval (−c, c),
then an P
= bn for everyPn. (Hint: Let
∞
f (x) = n=0 an x = ∞
n n
n=0 bn x . Differentiate term by term to show
(n)
that an and bn both equal f (0)/(n!).)
2. Show that if ∞ n
P
n=0 an x = 0 for all x in an open interval (−c, c),
then an = 0 for every n.
Exercise
Exercise 54 (The sum of the series).
P∞ 2 /2n )
n=0 (n To find the sum of this series, express 1/(1 − x) as
geometric series, differentiate both sides of the resulting equation with
respect ot x, multiply both sides of the result by x, differentiate again,
multiply by x again, and set x equal to 1/2. What do you get? (Source:
David E. Dobbs’ letter to the editor, Illinois Mathematics Teacher, Vol.33,
Issue 4, 1982,p.27.)
Exercise
Exercise 55 (Convergence at endpoints).
Show by examples that the convergence of a power series at an endpoint
of its interval of convergence, may be either conditional or absolute.
Exercise
Exercise 56.
Make up a power series whose interval of convergence is
(a) (-3,3)
(b) (-2,0)
(c) (1,5).
Taylor and Maclaurin Series
We now see how functions that are infinitely differentiable generate

power series called Taylor series.
In many cases, these series can provide useful polynomial

approximations of the generating functions.
Series Representations
We know that within its interval of convergence the sum of a power series
is a continuous function with derivatives of all orders.
But what about the other way around?
If a function f (x) has derivatives of all orders on an interval I , can it

be expressed as a power series on I ?
And if it can, what will its coefficients be?
We can answer the last question readily if we assume that f (x) is the sum
of a power series
∞
X
f (x) = an (x − a)n
n=0
= a0 + a1 (x − a) + a2 (x − a)2 + · · · + an (x − a)n + · · ·
with a positive radius of convergence.
By repeated term-by-term differentiation within the interval of
convergence I we obtain
f 0 (x) = a1 + 2a2 (x − a) + 3a3 (x − a)2 + · · · + nan (x − a)n−1 + · · ·
00
f (x) = 1.2a2 + 2.3a3 (x − a) + 3.4a4 (x − a)2 + · · ·
000
f (x) = 1.2.3a3 + 2.3.4a4 (x − a) + 3.4.5a5 (x − a)2 + · · ·
With the nth derivate, for all n, being

f (n) (x) = n!an + a sum of terms with (x − a) as a factor.
Since these equations all hold at x = a,we have
f 0 (a) = a1
f 00 (a) = 1.2a2
f 000 (a) = 1.2.3a3 .
..
.
f (n) (a) = n!an .
These formulas reveal a pattern in the coefficient of any power series

∞
X
an (x − a)n
n=0
that converges to the values of f on I (“represents f on I ”). If there is

such series (still an open question), then there is only one such series and
its nth coefficient is
f (n) (a)
an = .
n!
If f has a series representation, then the series must be
f 00 (a) f (n) (a)

f (x) = f (a)+f 0 (a)(x −a)+ (x −a2 )+· · ·+ (x −a)n +· · · (7)
2! n!
But if we start with an arbitrary function f that is infinitely differentiable

on an interval I centered at x = a and use it to generate the series in
Equation (7), will the series then converge to f (x) at each x in the interior
of I ?
The answer is maybe – for some functions it will but for other functions it
will not, as we will see.
Definition 57.
Let f be a function with derivatives of all orders throughout some interval
containing a as an interior point. Then the Taylor series generated by f at
x = a is
∞
X f (k) (a) f 00 (a) f (n) (a)
(x − a)k = f (a) + f 0 (a)(x − a) + (x − a)2 + · · · + (x − a)n + · · ·
k=0
k! 2! n!
The Maclaurin series generated by f is

∞
X f (k) (0) f 00 (0) 2 f (n) (0) n
x k = f (0) + f 0 (0)x + x + ··· + x + ··· ,
k! 2! n!
k=0
the Taylor series generated by f at x = 0.
The Maclaurin series generated by f is often just called the Taylor series of
f.
Finding a Taylor Series
Example 58.
Find the Taylor series generated by f (x) = 1/x at a = 2. Where, if
anywhere, does the series converge to 1/x?
We need to find f (2), f 0 (2), f 00 (2), · · · . Taking derivatives we get

f (x) = x −1 , f (x) = 2−1 = 21 ,
f 0 (x) = −x −2 , f 0 (2) = − 212 ,
f 00 (2) 1
f 00 (x) = 2!x −3 , 3!
= 2−3 = 23
,
f 000 (2)
f 000 (x) = −3!x −4 , 3!
= − 214 ,
.. ..
. .
f (n) (2) (−1)n

f n) (x) = (−1)n n!x −(n+1) , n!
= 2n+1
.
Finding a Taylor Series
The Taylor series is

f 00 (2) f (n) (2)
f (2) + f 0(2)(x − 2) + 2! (x − 2)2 + · · · + n! (x − 2)2 + · · ·
(x−2) (x−2)2 n
= 1
2 − 22
+ 23
− · · · + (−1)n (x−2)
2n+1
+ ··· .
This is a geometric series with first term 1/2 and ratio r = −(x − 2)/2. It
converges absolutely for |x − 2| < 2 and its sum is
1/2 1 1
= = .
1 + (x − 2)/2 2 + (x − 2) x
In this example the Taylor series generated by f (x) = 1/x at a = 2

converges to 1/x for |x − 2| < 2 or 0 < x < 4.
Taylor Polynomials
The linearizion of a differentiable function f at a point a is the polynomial

of degree one given by
P1 (x) = f (a) + f 0 (a)(x − a).
If f has derivatives of higher order at a, then it has higher-order

polynomial approximations as well, one for each available derivative.
These polynomials are called the Taylor polynomials of f .
Taylor Polynomial of order n
Definition 59.
Let f be a function with derivatives of order k for k = 1, 2, . . . , N in some
interval containing a as an interior point. Then for any integer n from 0
through N, the Taylor polynomial of order n generated by f at x = a is
the polynomial
f 00 (a) f (k) (a) f (n) (a)

Pn (x) = f (a) + f 0 (a)(x − a) + (x − a)2 + · · · + (x − a)k + · · · + (x − a)n .
2! k! n!
Taylor Polynomial of order n
We speak of a Taylor polynomial of order n rather than degree n because

f (n) (a) may be zero.
The first two Taylor polynomials of f (x) = cos x at x = 0, for example, are
P0 (x) = 1 and P1 (x) = 1.
The first-order Taylor polynomial has degree zero, not one.
Just as the linearization of f at x = a provides the best linear

approximation of f in the neighborhood of a, the higher-order Taylor
polynomials provide the best polynomial approximations of thier respective
degrees.
Example 60 (Finding Taylor Polynomials for e x ).
Find the Taylor series the Taylor polynomials generated by f (x) = e x at
x = 0.
Solution : Since f (x) = e x , f 0 (x) = e x , . . . , f (n) (x) = e x , . . . , we

have f (0) = e 0 = 1, f 0 (0) = 1, . . . , f (n) (0) = 1, . . ..
The Taylor series generated by f at x = 0 is
f 00 (0) f (n) (0) n x2 xn
f (0) + f 0 (0)x + + ··· + x + ··· = 1+x + + ··· + + ···
2! n! 2 n!
∞
X xk
= .
k=0
k!
This is also the Maclaurin series for e x . We will later see that the series
converges to e x at every x.
The Taylor polynomial of order n at x = 0 is
x2 xn
Pn (x) = 1 + x + + ··· + .
2 n!
Finding Taylor Polynomials for cos x
Example 61.
Find the Taylor series and Taylor polynomials generated by f (x) = cos x at
x = 0.
Solution : The cosine and its derivatives are
f (x) = cos x, f 0 (x) = − sin x,
f 00 (x) = − cos x, f (3) (x) = sin x,
f (2n) (x) = (−1)n cos x, f (2n+1) (x) = (−1)n+1 sin x.
At x = 0, the cosines are 1 and the sines are 0, so
f (2n) (0) = (−1)n , f (2n+1) (0) = 0.
The Taylor series generated by f at 0 is

f 00 (0) 2 f 000 (0) 3 f (n) (0) n
f (0) + f 0 (0)x + 2! x + 3! x + ··· + n! x + ···
x2 x4 x 2n
= 1 + 0.x − 2! + 0.x 3 + 4! + · · · + (−1)n (2n)! + ···
∞
X (−1)k x 2k
= !
(2k)
k=0
This is also the Maclaurin series for cos x. We will see that the series
converges to cos x at every x.
Because f (2n+1) (0) = 0, the Taylor polynomials of orders 2n and 2n + 1

are identical:
x2 x4 x 2n
P2n (x) = P2n+1 (x) = 1 − 2! + 4! − · · · + (−1)n (2n)! .
The following figure shows how well these polynomials approximate

f (x) = cos x near x = 0. Only the right-hand portions of the graphs are
given because the graphs are symmetric about the y-axis.
A Function f Whose Taylor Series Converges at Every x
but Converges to f (x) Only at x = 0
The following example shows that there is a function f whose Taylor series
converges at every x but converges to f (x) only at x = 0.
Example 62.
It can be shown (though not easily) that
(
0, x =0
f (x) = −1/x 2
e , x=6 0
has derivatives of all orders at x = 0 and that f (n) (0) = 0 for all n.
A function f whose Taylor series converges at every x but
converges to f (x) only at x = 0.
This means that the Taylor series generated by f at x = 0 is
f 00 (0) 2 f (n) (0) n
f (0) + f 0 (0)x + x + ··· + x + · · · = 0 + 0.x + 0.x 2 + · · · + 0.x n + · · · = 0.
2! n!
The series converges for every x (its sum is 0) but converges to f (x) only
at x = 0.
Some questions
Two questions still remain
1. For what values of x can we normally expect a Taylor series to

converge to its generating function?
2. How accurately do a function’s Taylor polynomials approximate the
function on a given interval?
The answers are provided by a theorem of Taylor, which will be discussed

next.
Finding Taylor Polynomials
Exercise 63.
In the following exercises, find the Taylor polynomials of orders 0, 1, 2, and
3 generated by f at a.
1. f (x) = ln x. a=1
2. f (x) = 1/(x + 2), a = 0
3. f (x) = sin x, a = π/4
√
4. f (x) = x, a=4
√
5. f (x) = x + 4, a=0
Finding Taylor Series at x=0 (Maclaurin Series)
Exercise 64.
Find the Maclaurin series for the function in the following exercises.
1. e −x
1
2. 1−x
3. sin x2
4. 7 cos(−x)
e x −e −x
5. sinh x = 2
6. x 4 − 2x 3 − 5x + 4
Finding Taylor Series
Exercise 65.
In the following exercises, find the Taylor series generated by f at x = 0.
1. f (x) = x 3 − 2x + 4, a=2
2. f (x) = x4 + x2 + 1, a = −2
3. f (x) = 3x 5 − x4 + 2x 3 + x 2 − 2, a = −1
4. f (x) = x/(1 − x), a=0
5. f (x) = ex , a=2
6. f (x) = 2x , a=1
Exercise
Exercise 66.
(a) Use the Taylor series generated by e x at x = a to show that
h (x − a)2 i
e x = e a 1 + (x − a) + + ··· .
2!
(b) Find the Taylor series generated by e x at x = 1. Compare your
answer with the formula in the above exercise.
(c) Let f (x) have derivatives through order n at x = a. Show that the
Taylor polynomial of order n and its first n derivatives have the same
values that f and its first n derivatives have at x = a.
Of all polynomials of degree ≤ n, the Taylor polynomial of
order n gives the best approximation
Exercise 67.
Suppose that f (x) is differentiable on an interval centered at x = a and
that g (x) = b0 + b1 (x − a) + · · · + bn (x − a)n is a polynomial of degree n
with constant coefficient b0 , . . . , bn . Let E (x) = f (x) − g (x). Show that if
we impose on g conditions
(a) E (a) = 0 (the approximation error is zero at x = a )
E (x)
(b) limx→a (x−a)n = 0 (the error is negligible when compare to (x − a)n )
then
f 00 (a) f (n) (a)

g (x) = f (a) + f 0 (a)(x − a) + (x − a)2 + · · · + (x − a)n .
2! n!
Thus, the Taylor polynomial Pn (x) is the only polynomial of degree less
than or equal to n whose error is both zero at x = a and negligible when
compared with (x − a)n .
Quadratic Approximations
Exercise 68.
The Taylor polynomial of order 2 generated by a twice-differentiable
function f (x) at x = a is called the quadratic approximation of f at x = a.
Find the
(a) linearizion (Taylor polynomial of order 1) and
(b) quadratic approximation of f at x = 0.
in the following exercises.
1. f (x) = ln(cos x)
2. f (x) = e sin x
√
3. f (x) = 1/ 1 − x 2
4. f (x) = cosh x
5. f (x) = sin x
6. f (x) = tan x
Convergence of Taylor Series
Convergence of Taylor Series
We now address the following two questions.
1. When does a Taylor series converge to its generating function?

2. How accurately do a function’s Taylor polynomials approximate the
function on a given interval?
Taylor’s Theorem
We answer the questions with the following theorem.

Theorem 69.
If f and its first n derivatives f 0 , f 00 , . . . , f (n) are continuous on the closed
interval between a and b, and f (n) is differentiable on the open interval
between a and b, then there exists a number c between a and b such that
f 00 (a) f (n) (a) f (n+1) (c)

f (b) = f (a) + f 0 (a) (b − a) + (b − a)2 + · · · + (b − a)n + (b − a)n+1 .
2! n! (n + 1)!
Taylor’s Theorem is a generalization of the Mean Value Theorem.
Proof of the theorem will be discussed at the end of this lecture.
Taylor’s Formula
When we apply Taylor’s Theorem, we usually want to hold “a” fixed and
treat “b” as an independent variable. Taylor’s formula is easier to use in
circumstances like these if we change b to x. Here is a version of the
theorem with this change.
Theorem 70.
If f has derivatives of all orders in an open interval I containing a, then for
each positive integer n and for each x in I .
f 00 (a) f (n) (a)
f (x) = f (a) + f 0 (a) (x − a) + (x − a)2 + · · · + (x − a)n + Rn (x) , (8)
2! n!
where
f (n+1) (c)
Rn (x) = (x − a)n+1 (9)
(n + 1)!
for some c between a and x.
Taylor’s Formula
When we state Taylor’s theorem this way, it says that for each x ∈ I ,
f (x) = Pn (x) + Rn (x).
The function Rn (x) is determined by the value of the (n + 1)st derivative

f (n+1) at a point c that depends on both a and x, and that lies somewhere
between them.
For any value of n we want, the equation gives both a polynomial

approximation of f of that order and a formula for the error involved in
using that approximation over the interval I .
Taylor’s Formula
Equation (8) is called Taylor’s formula. The function Rn (x) is called the
remainder of order n or the error term for the approximation of f by
Pn (x) over I .
If Rn (x) → 0 as n → ∞ for all x ∈ I , we say that the Taylor series

generated by f at x = a converges to f on I , and we write
∞
X f (k) (a)
f (x) = (x − a)k .
k!
k=0
Often we can estimate Rn without knowing the value of c, as the following

example illustrates.
Example
Example 71.
Show that the Taylor series generated by f (x) = e x at x = 0 converges to
f (x) for every real value of x.
Solution. The function has derivatives of all orders throughout the interval
I = (−∞, ∞). Equations (8) and (9) with f (x) = e x and a = 0 give
x2 xn
ex = 1 + x + + ··· + + Rn (x)
2! n!
and
ec
Rn (x) = x n+1 for some c between 0 and x.
(n + 1)!
Since e x is an increasing function of x, e c lies between e 0 = 1 and e x .
When x is negative, so is c, and e c < 1. When x is zero, e x = 1 and
Rn (x) = 0. When x is positive, so is c, and e c < e x .
Example
Thus, for Rn (x) given as above,

|x|n+1
|Rn (x)| ≤ when x ≤ 0, ec < 1
(n + 1)!
and
x n+1
|Rn (x)| < e x when x > 0. ec < ex
(n + 1)!
Finally, because
x n+1
lim =0 for every x,
n→∞ (n + 1)!
lim Rn (x) = 0, and the series converges to e x for every x. Thus,

n→∞
∞
X xk x2 xk
ex = =1+x + + ··· + + ··· . (10)
k! 2! k!
k=0
Example
We can use the result of Example 71 with x = 1 to write

1 1
e =1+1+ + ··· + + Rn (1),
2! n!
where for some c between 0 and 1,
1 3
Rn (1) = e c < since e c < e 1 < 3.
(n + 1)! (n + 1)!
Estimating the Remainder
It is often possible to estimate Rn (x) as we did in Example 71. This

method of estimation is so convenient that we state it as a theorem for
future reference.
Theorem 72 (The Remainder Estimation Theorem).
If there is a positive constant M such that |f (n+1) (t)| ≤ M for all t
between x and a, inclusive, then the remainder term Rn (x) in Taylor’s
Theorem satisfies the inequality
|x − a|n+1
|Rn (x)| ≤ M .
(n + 1)!
If this inequality holds for every n and the other conditions of Taylor’s
Theorem are satisfied by f , then the series converges to f (x).
Exercise
We are now ready to look at some examples of how the Remainder
Estimation Theorem and Taylor’s Theorem can be used together to settle
questions of convergence. As we will see, they can also be used to
determine the accuracy with which a function is approximated by one of
its Taylor polynomials.
Example 73.
Show that the Taylor series for sin x at x = 0 converges for all x.
Solution. The function and its derivatives are

f (x) = sin x, f 0 (x) = cos x,
00 000
f (x) = − sin x, f (x) = − cos x,
.. ..
. .
f (2k) (x) = (−1)k sin x, f (2k+1) (x) = (−1)k cos x,
so
f (2k) (0) = 0 and f (2k+1) (0) = (−1)k .
Solution (contd...)
The series has only odd-powered terms and, for n = 2k + 1, Taylor’s Theorem gives
x3 x5 (−1)k x 2k+1
sin x = x − + − ··· + + R2k+1 (x).
3! 5! (2k + 1)!
All the derivatives of sin x have absolute values less than or equal to 1, so we can apply the
Remainder Estimation Theorem with M = 1 to obtain
|x|2k+2
|R2k+1 (x)| ≤ 1 · .
(2k + 2)!
Since (|x|2k+2 /(2k + 2)!) → 0 as k → ∞, whatever the value of x, so R2k+1 (x) → 0 and the
Maclaurin series for sin x converges to sin x for every x. Thus,
∞
X (−1)k x 2k+1 x3 x5 x7
sin x = =x− + − + ··· .
k=0
(2k + 1)! 3! 5! 7!
Example 74.
Show that the Taylor series for cos x at x = 0 converges to cos x for every
value of x.
Solution. We add the remainder term to the Taylor polynomial for cos x
to obtain Taylor’s formula for cos x with n = 2k;
x2 x4 x 2k
cos x = 1 − + − · · · + (−1)k · · · + R2k (x) .
2! 4! (2k)!
Because the derivatives of the cosine have absolute value less than or
equal to 1, the Remainder Estimation Theorem with M = 1 gives
|x|2k+1
|R2k (x) | ≤ 1. .
(2k + 1)!
For every value of x, R2k (x) → 0 as k → ∞. Therefore, the series
converges to cos x for every value of x. Thus,
∞
X (−1)k x 2k x2 x4 x6
cos x = =1− + − + ··· . (11)
(2k)! 2! 4! 6!
k=0
Using Taylor Series
Since every Taylor series is a power series, the operations of adding,

subtracting, and multiplying Taylor series are all valid on the intersection
of their intervals of convergence.
Example 75.
Using known series, find the first few terms of the Taylor series for the
given function using power series operations.
1
(a) 3 (2x + x cos x)
(b) ex cos x
Solution
x2 x4 x 2k

1 2 1
(a) (2x + x cos x) = x + x 1− + − · · · + (−1)k + ···
3 3 3 2! 4! (2k)!
2 1 x3 x5 x3 x5
= x+ x− + − ··· − x − + − ···
3 3 3! 3.4! 6 72
x2 x3 x4 x2 x4

(b) e x cos x = 1 + x + + + + ··· . 1− + − ···
2! 3! 4! 2! 4!
x2 x3 x4
2
x3 x4 x5

x
= 1+x + + + + ··· − + + + ···
2! 3! 4! 2! 2! 2!2! 2!3!
4
x5 x6

x
+ + + + ··· + ···
4! 4! 2!4!
x3 x4
=1 + x − − + ···
3 6
Using Taylor series
P∞ n
We recall that if n=0 an x converges absolutely for |x| < R, then
∞
X
an (f (x))n
n=0
converges absolutely for any continuous function f on |f (x)| < R.
We can use the Taylor series of the function f to find the Taylor series of
f (u (x)) where u (x) is any continuous function.
The Taylor series resulting from this substitution will converge for all x
such that u (x) lies within the interval of convergence of the Taylor series
of f .
Using Taylor series
For instance, we can find the Taylor series for cos 2x by substituting 2x
for x in the Taylor series for cos x :
∞
X (−1)k (2x)2k (2x)2 (2x)4 (2x)6
cos 2x = =1− + − + ···
(2k)! 2! 4! 6!
k=0
22 x 2 24 x 4 26 x 6
=1 − + − + ···
2! 4! 6!
∞
X 22k x 2k
= (−1)k .
(2k)!
k=0
Example 76.
For what values of x can we replace sin x by x − x 3 /3! with an error of

magnitude no greater than 3 × 10−4 ?
Solution. Here we can take advantage of the fact that the Taylor series
for sin x is an alternating series for every nonzero value of x. According to
the Alternating Series Estimation Theorem, the error in truncating
x3 x5
sin x = x − + − ···
3! 5!
after x 3 /3! is no greater than

5
x |x|5
=
5! 120 .
Therefore the error will be less than or equal to 3 × 10−4 if
|x|5 p
< 3 × 10−4
5
or |x| < 306 × 10−4 ≈ 0.514.
120
The Alternating Series Estimation Theorem tells us something that the

Remainder Estimation Theorem does not: namely, that the estimate
x − x 3 /3! for sin x is an underestimate when x is positive, because then
x 5 /120 is positive.
The above figure shows the graph of sin x, along with the graphs of a
number of its approximating Taylor polynomials.
The graph of P3 (x) = x − x 3 /3! is almost indistinguishable from the

sine curve when 0 ≤ x ≤ 1.
Proof of Taylor’s Theorem
We prove Taylor’s theorem assuming a < b. The proof for a > b is nearly
the same. The Taylor polynomial
f 00 (a) f (n) (a)

Pn (x) = f (a) + f 0 (a) (x − a) + (x − a)2 + · · · + (x − a)n
2! n!
and its first n derivatives match the function f and its first n derivatives at
x = a. We do not disturb that matching if we add another term of the
form K (x − a)n+1 , where K is any constant, because such a term and its
first n derivatives are all equal to zero at x = a. The new function
φn (x) = Pn (x) + K (x − a)n+1
and its first n derivatives still agree with f and its first n derivatives at
x = a.
Proof of Taylor’s Theorem (contd...)
We now choose the particular value of K that makes the curve y = φn (x)
agree with the original curve y = f (x) at x = b. In symbols,
f (b) − Pn (b)
f (b) = Pn (b) + K (b − a)n+1 , or K= . (12)
(b − a)n+1
With K defined by Equation (12), the function
F (x) = f (x) − φn (x)
measures the difference between the original function f and the

approximating function φn for each x in [a, b].
We now use Rolle’s Theorem. First, because F (a) = F (b) = 0 and both
F and F 0 are continuous on [a, b] , we know that
F 0 (c1 ) = 0
for some c1 in (a, b) . Next, because F 0 (a) = F 0 (c1 ) = 0 and both F 0 and
F 00 are continuous on [a, c1 ] we know that
F 00 (c2 ) = 0
for some c2 in (a, c1 ) . Rolle’s Theorem, applied successively to
F 00 , F 000 , . . . , F (n−1) implies the existence of
c3 in (a, c2 ) such that F 000 (c3 ) = 0,
c4 in (a, c3 ) such that F (4) (c4 ) = 0,
..
.
cn in (a, cn−1 ) such that F (n) (cn ) = 0.
Finally, because F (n) is continuous on [a, cn ] and differentiable on (a, cn ),

and F (n) (a) = F (n) (cn ) = 0, Rolle’s Theorem implies that there is a
number cn+1 in (a, cn ) such that
F (n+1) (cn+1 ) = 0 (13)
If we differentiate F (x) = f (x) − Pn (x) − K (x − a)n+1 a total of n + 1

times, we get
F (n+1) (x) = f (n+1) (x) − 0 − (n + 1)!K . (14)
Equations (13) and (14) together give
f (n+1) (c)
K= for some number c = cn+1 in (a, b) . (15)
(n + 1)!
Equation (12) and (15) give
f (n+1) (c)
f (b) = Pn (b) + (b − a)n+1 .
(n + 1)!
This concludes the proof.
Finding Taylor Series
Exercises 77.
Use substitution to find the Taylor series at x = 0 of the functions in the
following exercises.
1. sin πx

2
√
2. cos x 2/3 / 2
3. tan−1 (3x 4 )
1
4. 1+ 34 x 3
Exercises
Exercises 78.
Use power series operations to find the Taylor series at x = 0 for the
functions in the following exercises.
1. xe x
x2
2. 2 − 1 + cos x
3. x ln(1 + 2x)
4. sin x · cos x
5. cos x − sin x
6. ln(1 + x) − ln(1 − x)
Exercises
Exercises 79.
Find the first four nonzero terms in the Maclaurin series for the functions
in the following exercises.
ln(1+x)
1. 1−x
2. (tan−1 x)2
3. cos2 x · sin x
4. sin(tan−1 x)
Error Estimates
Exercises 80.
1. Estimate the error if P3 (x) = x − (x 3 /6) is used to estimate the value
of sin x at x = 0.1.
2. Estimate the error if P4 (x) = 1 + x + (x 2 /2) + (x 3 /6) + (x 4 /24) is
used to estimate the value of e x at x = 1/2.
3. For approximately what values of x can you replace sin x by
x − (x 3 /6) with an error of magnitude no greater than 5 × 10−4 ?
Give reasons for your answer.
4. If cos x is replaced by 1 − (x 2 /2) and |x| < 0.5, what estimate can be
made of the error? Does 1 − (x 2 /2) tend to be too large, or too
small? Give reasons for your answer.
Exercises
Exercises 81.
1. How close is the approximation sin x = x when |x| < 10−3 ? For
which of these values of x is x < sin x?
√
2. The estimate 1 + x = 1 + (x/2) is used when x is small. Estimate
the error when |x| < 0.01.
3. The approximation e x = 1 + x + (x 2 /2) is used when x is small. Use
the Remainder Estimation Theorem to estimate the error when
|x| < 0.1.
4. (Continuation of the above exercise) When x < 0, the series for e x is
an alternating series. Use the Alternating Series Estimation Theorem
to estimate the error that results from replacing e x by 1 + x + (x 2 /2)
when −0.1 < x < 0. Compare your estimate with the one you
obtained in the above exercise.
Theory and Examples
Exercises 82.
1. Use the identity sin2 x = (1 − cos 2x)/2 to obtain the Maclaurin series
for sin2 x. Then differentiate this series to obtain the Maclaurin series
for 2 sin x cos x. Check that this is the series for sin 2x.
2. (Continuation of the above xercise.) Use the identity
cos2 x = cos 2x + sin2 x to obtain a power series for cos2 x.
3. Taylor’s Theorem and the Mean Value Theorem. Explain how the
Mean Value Theorem is a special case of Taylor’s Theorem.
4. Linearizations at inflection points. Show that if the graph of a
twice-differentiable function f (x) has an inflection point at x = a,
then the linearization of f at x = a is also the quadratic
approximation of f at x = a. This explains why tangent lines fit so
well at inflection points.
Theory and Examples
Exercises 83.
1. The (second) second derivative test. Use the equation
f 00 (c2 )
f (x) = f (a) + f 0 (a)(x − a) + (x − a)2
2
to establish the following test:
Let f have continuous first and second derivatives and suppose that
f 0 (a) = 0. Then
a. f has a local maximum at a if f 00 ≤ 0 throughout an interval whose
interior contains a;
b. f has a local maximum at a if f 00 ≥ 0 throughout an interval whose
interior contains a.
Exercises
Exercises 84.
1. A cubic approximation. Use Taylor’s formula with a = 0 and n = 3 to
find the standard cubic approximation of f (x) = 1/(1 − x) at x = 0.
Give an upper bound for the magnitude of the error in the
approximation when |x| ≤ 0.1.
2. a. Use Taylor’s formula with n = 2 to find the quadratic approximation of
f (x) = (1 + x)k at x = 0 (k a constant).
b. If k = 3, for approximately what values of x in the interval [0, 1] will
the error in the quadratic approximation be less than 1/100?
3. Improving approximations of π:
a. Let P be an approximation of π accurate to n decimals. Show that
P + sin P gives an approximation correct to 3n decimals. (Hint : Let
P = π + x.)
b. Try it with a calculator.
P∞ n
The Taylor series generated by f (x) = n=0 an x .
Exercise 85.
A function defined by a power series ∞ n
P
n=0 an x with a radius of
convergence R > 0 has a Taylor series that converges to the function at
every point of (−R, R).
P Show this by showingPthat the Taylor series
generated by f (x) = ∞ n=0 an x n is the series ∞ n
n=0 an x itself.
An immediate consequence of this is that series like
x4 x6 x8
x sin x = x 2 − + − + ...
3! 5! 7!
and
x4 x5
x 2e x = x 2 + x 3 +
+ + ...,
2! 3!
obtained by multiplying Taylor series by powers of x, as well as series
obtained by integration and differentiation of convergent power series, are
themselves the Taylor series generated by the functions they represent.
Exercises
Exercises 86.
1. Taylor series
P for even functions and odd functions. Suppose that
f (x) = ∞ n=0 an x n converges for all x in an open interval (−R, R).
Show that
a. If f is even, then a1 = a3 = a5 = · · · = 0, i.e., the Taylor series for f at
x =0 contains only even powers of x.
b. If f is odd, then a0 = a2 = a4 = · · · = 0, i.e., the Taylor series for f at
x =0 contains only odd powers of x.
References
1. M.D. Weir, J. Hass and F.R. Giordano, Thomas’ Calculus, 11th

Edition, Pearson Publishers.
2. N. Piskunov, Differential and Integral Calculus, Vol I & II (Translated
by George Yankovsky).
3. S.C. Malik and Savitha Arora, Mathematical Analysis, New Age
Publishers.
4. R. G. Bartle, D. R. Sherbert, Introduction to Real Analysis, Wiley
Publishers.

Infinite Series Part 2

Uploaded by

Copyright:

Available Formats

Infinite Series Part 2

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Infinite Series Part 2

Uploaded by

Copyright:

Available Formats

Infinite Series

We now discuss convergence of series whose terms are alternately positive

Series (1), called the alternating harmonic series, converges, as we will

We prove the convergence of the alternating harmonic series by applying

converges if all three of the following conditions are satisfied:

s2m = (u1 − u2 ) + (u3 − u4 ) + · · · + (u2m−1 − u2m )

Hence s2m+2 ≥ s2m , and the sequence {s2m } is non-decreasing and

If n is an odd integer, say n = 2m + 1, then the sum of the first n terms is

s2m+1 = s2m + u2m+1 → L + 0 = L.

We have the following result for sequences :

As s2m+1 → L and s2m → L, limn→∞ sn = L.

satisfies the three requirements of Leibniz’s Theorem with N = 1; it

Then from the Derivative Quotient Rule,

It follows that un ≥ un+1 for n ≥ 4. That is, the sequence {un } is

A graphical interpretation of the partial sums shows how an alternating

To find the point corresponding to s2 = u1 − u2 , we back up a distance

We continue in this seesaw fashion, backing up or going forward as the

We oscillate across the limit L, and the amplitude of oscillation approaches

Because |L − sn | < un+1 for n ≥ N,

We can make useful estimates of the sums of convergent alternating series.

satisfies the three conditions of Leibniz’s Theorem, then for n ≥ N,

The Alternating Series Estimation Theorem says that if we truncate the

1/2n . The original

The alternating harmonic series is conditionally convergent, or converges

If p > 1, the series converges absolutely. If 0 < p ≤ 1, the series converges

Conditional convergence: 1 − √1 + √1 − √1 + ···.

We need to be careful when using a conditionally convergent series. We

We can always rearrange the terms of a finite collection of numbers

Absolutely convergence is important for two reasons. First, we have good

On the other hand, if we rearrange the terms of a conditionally convergent

Here is an example of summing the terms of a conditionally convergent

So by rearranging the terms of the conditionally convergent series

the series becomes

Example 11 shows that we cannot rearrange the terms of a conditionally

When we use a conditionally convergent series, the terms must be added

In contrast, Theorem 10 guarantees that the terms of an absolutely

Let ε be a positive real number, let L = ∞

Pnin the sequence {bn },

| nk=1 bk − L| ≤ | nn=1 bk − sN2 | + |sN2 − L|

The argument in part (a)Pshows thatPif ∞

Now show that because ∞

A possible rearrangement of the terms of the series might start with a

In this example, if we had the second series to begin with, we would

We can do even better: The sum of either series is also equal to

If we rearrange infinitely many terms of conditionally convergent series, we

Example 14 (Rearranging the Alternating Harmonic Seires).

This process can be continued indefinitely. Because both the

We have now developed several tests for convergence and divergence of

1. The nth-Term Test: Unless an → 0, the series diverges.

with an error of magnitude less than 5 × 10−6 .

converges absolutely, then

converge or diverge? Justify your answer.

Like polynomials, power series can be added, subtracted, multiplied,

A power series about x = 0 is a series of the form

A power series about x = a is a series of the form

in which the center a and the coefficients c0 , c1 , c2 , . . . , cn , . . . are

Equation (7) is the special case obtained by taking a = 0 in Equation (2).

This is the geometric series with first term 1 and ratio x.

It converges to 1/(1 − x) for |x| < 1. We express this fact by writing

We think of the partial sums of the series on the right as polynomials