MATH 111 Calculus 2, NUU, Spring 2026

Dr. K. Mamayusupov Tutorial 5

Problem 1
If f (n) (0) = (n + 1)! for n = 0, 1, 2, . . . , find the Maclaurin series for f and its radius
of convergence.

Solution:
∞ ∞ ∞
X f (n) (0) X (n + 1)! X
xn = xn = (n + 1)xn .
n=0
n! n=0
n! n=0

(Ratio Test.) Here an = (n + 1)xn . Then

an+1 (n + 2)xn+1 n+2

= = |x| −−−→ |x|.
an (n + 1)xn n+1 n→∞

So the series converges if |x| < 1 and diverges if |x| > 1, hence R = 1. Note that
∞
X 1
(n + 1)xn = (|x| < 1).
n=0
(1 − x)2

Problem 2
Use the definition of a Taylor series to find the first four nonzero terms of the series
for f (x) centered at the given value of a
x
1) f (x) = xe
√ , a=0
2) f (x) = 3 x, a = 8
π
3) f (x) = sin x, a =
6

Solution:
(1) f (x) = xex , a = 0. Using the Taylor formula

f ′′ (0) 2 f ′′′ (0) 3 f (4) (0) 4

f (0) + f ′ (0)x + x + x + x + ··· ,
2! 3! 4!
compute derivatives:

f (x) = xex , f ′ (x) = ex (x + 1), f ′′ (x) = ex (x + 2), f ′′′ (x) = ex (x + 3),

f (4) (x) = ex (x + 4).

Hence f (0) = 0, f ′ (0) = 1, f ′′ (0) = 2, f ′′′ (0) = 3, f (4) (0) = 4, and

x3 x4
x + x2 + + + ··· .
2 6

Page 2
 (2) f (x) = x1/3 , a = 8. Let h = x − 8. The Taylor series at a = 8 is

f ′′ (8) 2 f ′′′ (8) 3

f (8) + f ′ (8)h + h + h + ··· .
2! 3!
Compute derivatives:
1 2 10 −8/3
f ′ (x) = x−2/3 , f ′′ (x) = − x−5/3 , f ′′′ (x) = x .
3 9 27
At x = 8: 81/3 = 2, 8−2/3 = 41 , 8−5/3 = 1
32
, 8−8/3 = 1
256
, so
1 1 5
f (8) = 2, f ′ (8) = , f ′′ (8) = − , f ′′′ (8) = .
12 144 3456
Therefore
1 1 5
2+ (x − 8) − (x − 8)2 + (x − 8)3 + · · · .
12 288 20736

(3) f (x) = sin x, a = π6 . Using the Taylor formula

′′ π  ′′′ π 



π  π   π  f π 2 f π 3
f + f′ x− + 6
x− + 6
x− + ··· ,
6 6 6 2! 6 3! 6
and the derivatives f ′ (x) = cos x, f ′′ (x) = − sin x, f ′′′ (x) = − cos x, with
π  1  π  √3 π  1 π  √
3
f = , f′ = , f ′′ = − , f ′′′ =− ,
6 2 6 2 6 2 6 2
we obtain the first four nonzero terms:
√  √ 
1 3 π 1  π 2 3 π 3
+ x− − x− − x− + ··· .
2 2 6 4 6 12 6

Problem 3
Find the Maclaurin series for f (x) using the definition of a Maclaurin series. [Assume
that f has a power series expansion. Do not show that Rn (x) → 0.] Also find the
associated radius of convergence.

1) f (x) = (1 − x)−2 2)f (x) = cos x 3) f (x) = sinh x

Solution:
(1) f (x) = (1 − x)−2 . Start with the geometric series
∞
1 X
= xn (|x| < 1).
1 − x n=0

Page 3
 Differentiate term-by-term:
∞
  ! ∞ ∞
d 1 1 d X
n
X
n−1
X
= and x = nx = (n + 1)xn .
dx 1 − x (1 − x)2 dx n=0 n=1 n=0

Therefore,
∞
1 X
= (n + 1)xn (|x| < 1),
(1 − x)2 n=0
so the radius of convergence is R = 1. (Also by derivatives.)
Since f (x) = (1 − x)−2 , we have
f (n) (x) = (n + 1)!(1 − x)−(n+2) ⇒ f (n) (0) = (n + 1)!,
so ∞ ∞
X f (n) (0) X
n
f (x) = x = (n + 1)xn ,
n=0
n! n=0
with the same radius of convergence R = 1.
(2) f (x) = cos x. The derivatives cycle:
f (0) = 1, f ′ (x) = − sin x ⇒ f ′ (0) = 0, f ′′ (x) = − cos x ⇒ f ′′ (0) = −1,
f ′′′ (x) = sin x ⇒ f ′′′ (0) = 0, . . .
In general, f (2n) (0) = (−1)n and f (2n+1) (0) = 0, so
∞
X (−1)n
cos x = x2n .
n=0
(2n)!
Using the Ratio Test,
x2(n+1)
(2n+2)! |x|2
lim x2n
= lim = 0,
n→∞ n→∞ (2n + 2)(2n + 1)
(2n)!

so R = ∞.
(3) f (x) = sinh x. Using derivatives:
f (0) = 0, f ′ (x) = cosh x ⇒ f ′ (0) = 1, f ′′ (x) = sinh x ⇒ f ′′ (0) = 0,
f ′′′ (x) = cosh x ⇒ f ′′′ (0) = 1, . . .
In general, f (2n) (0) = 0 and f (2n+1) (0) = 1, so
∞
X x2n+1
sinh x = .
n=0
(2n + 1)!
Using the Ratio Test,
x2(n+1)+1
(2n+3)! |x|2
lim x2n+1
= lim = 0,
n→∞ n→∞ (2n + 3)(2n + 2)
(2n+1)!

so R = ∞.

Page 4
Problem 4
Find the Taylor series for f (x) centered at the given value of a. [Assume that f has
a power series expansion. Do not show that Rn (x) → 0.] Also find the associated
radius of convergence.

1) f (x) = x5 + 2x3 + x, a=2 2) f (x) = ln x, a=2

Solution:
(1) Compute derivatives directly and use the definition of the Taylor series at a = 2:
∞
X f (n) (2)
f (x) = (x − 2)n .
n=0
n!

We have
f ′ (x) = 5x4 +6x2 +1, f ′′ (x) = 20x3 +12x, f (3) (x) = 60x2 +12, f (4) (x) = 120x,
f (5) (x) = 120, f (n) (x) = 0 (n ≥ 6).
Evaluating at x = 2:
f (2) = 50, f ′ (2) = 105, f ′′ (2) = 184, f (3) (2) = 252, f (4) (2) = 240,
f (5) (2) = 120.
Therefore,
184 252 240 120
f (x) = 50 + 105(x − 2) + (x − 2)2 + (x − 2)3 + (x − 2)4 + (x − 2)5 ,
2! 3! 4! 5!
i.e.
f (x) = 50 + 105(x − 2) + 92(x − 2)2 + 42(x − 2)3 + 10(x − 2)4 + (x − 2)5 .

Since f is a polynomial, its Taylor series about a = 2 is finite (it equals the
polynomial).
Alternatively, let h = x − 2, so x = 2 + h.
f (x) = x5 + 2x3 + x = (2 + h)5 + 2(2 + h)3 + (2 + h).
Expand:
(2 + h)5 = 32 + 80h + 80h2 + 40h3 + 10h4 + h5 ,
2(2 + h)3 = 2(8 + 12h + 6h2 + h3 ) = 16 + 24h + 12h2 + 2h3 ,
and (2 + h) = 2 + h. Summing,
f (x) = 50 + 105h + 92h2 + 42h3 + 10h4 + h5 ,
i.e.
f (x) = 50 + 105(x − 2) + 92(x − 2)2 + 42(x − 2)3 + 10(x − 2)4 + (x − 2)5 .
The radius of convergence is R = ∞.

Page 5
(2) Let h = x − 2. Then x = 2 + h and
 
h
ln x = ln(2 + h) = ln 2 + ln 1 + .
2
P∞ n−1 un
Use ln(1 + u) = n=1 (−1) n
for |u| < 1 with u = h2 :
∞  n
X
n−1 1 x−2
ln x = ln 2 + (−1) .
n=1
n 2

x−2
This converges when 2
< 1, i.e. |x − 2| < 2, so R = 2.
(Alternatively, by computing derivatives directly.) Using the definition of the
Taylor series at a = 2,
∞
X f (n) (2)
ln x = (x − 2)n , f (x) = ln x.
n=0
n!

We have
1 1 1 1
f (2) = ln 2, f ′ (x) = ⇒ f ′ (2) = , f ′′ (x) = − 2
⇒ f ′′ (2) = − ,
x 2 x 4
2 1 6 3
f (3) (x) = 3
⇒ f (3) (2) = , f (4) (x) = − 4
⇒ f (4) (2) = − , . . .
x 4 x 8
In general, for n ≥ 1,
(n − 1)! (n − 1)!
f (n) (x) = (−1)n−1 ⇒ f (n) (2) = (−1)n−1 .
xn 2n
Therefore
∞ ∞  n
X
n−1 (n − 1)! n
X
n−1 1 x−2
ln x = ln 2 + (−1) (x − 2) = ln 2 + (−1) ,
n=1
2n n! n=1
n 2

with radius of convergence R = 2.

Page 6
Problem 5
Prove that the series obtained in Problem 3 (exercise 3) represents sinh x for all x.

Solution: Let ∞
X x2n+1
S(x) = .
n=0
(2n + 1)!
This is the Maclaurin series for sinh x that we obtained in Problem 3(3). We justify
that it represents sinh x for all real x using Taylor’s Inequality.
Since sinh x has derivatives
(sinh x)′ = cosh x, (sinh x)′′ = sinh x, (sinh x)(3) = cosh x, . . .
we have, for every n, that (sinh x)(n) (t) equals either sinh t or cosh t. Fix d > 0 and
assume |x| < d. Then for |t| ≤ d,
(sinh x)(n) (t) ≤ cosh(d).

Page 7
Let Tn (x) be the nth-degree Maclaurin polynomial for sinh x and Rn (x) = sinh x −
Tn (x) the remainder. By Taylor’s Inequality,
M
|Rn (x)| ≤ |x|n+1 ,
(n + 1)!

where M = max|t|≤d (sinh x)(n+1) (t) . We may take M = cosh(d), hence

cosh(d) n+1
|Rn (x)| ≤ |x| −−−→ 0.
(n + 1)! n→∞

Therefore the Maclaurin series of sinh x converges to sinh x for all x, i.e.
∞
X x2n+1
sinh x = for all x.
n=0
(2n + 1)!

Problem 6
Use the binominal series to expand the given function as a power series. State the
radius of convergence.
√
4 1
1) 1−x 2)
(2 + x)3

Solution:
(1) Use the binomial series
∞  
k
X k n
(1 + u) = u , |u| < 1,
n=0
n
1
with k = 4
and u = −x:
∞
√
1
X 1 3 7 3 77 4
4
1 − x = (1 − x)1/4 = 4(−x)n = 1 − x − x2 − x − x −··· .
n=0
n 4 32 128 2048

Binomial-coefficient calculations:
1
4 = 1,
0

Page 8
 1
4
1
= ,
1 4
1 1
1

− 1 3
4 = 4 4 =− ,
2 2! 32
1 1
− 1 14 − 2
1


7
4 = 4 4 = ,
3 3! 128
1 1
1
1
1

− 1 − 2 − 3 77
4 = 4 4 4 4
=− .
4 4! 2048
1
n 4
So (−1) gives the coefficients of xn .
n

Here |x| < 1, so R = 1.

(2) Rewrite
1 1 x −3
= 1 + .
(2 + x)3 8 2
Use the binomial series with k = −3 and u = x2 :
∞   ∞
 x −3 X −3  x n X n (n + 2)(n + 1) x
 n
1+ = = (−1) .
2 n=0
n 2 n=0
2 2

Binomial-coefficient calculations:
 
−3 (−3)(−4) · · · (−3 − n + 1) 3 · 4 · · · (n + 2) (n + 2)!
= = (−1)n = (−1)n
n n! n! 2! n!
=
(n + 2)(n + 1)
(−1)n .
2
Therefore
∞ ∞
1 X
n (n + 2)(n + 1) xn X (n + 2)(n + 1) n
3
= (−1) · n+3
= (−1)n x .
(2 + x) n=0
2 2 n=0
2n+4

x
Since 2
< 1, the radius of convergence is R = 2.

Page 9
Problem 8
Use a Maclaurin series from the table to obtain the Maclaurin series for the given
function.

1) f (x) = arctan (x2 )

2) f (x) = x cos 2x
x
3) f (x) = √
4 + x2
4) f (x) = sin2 x [Hint: Use sin2 x = 21 (1 − cos 2x).]

Solution:
(1) From the table,
∞
X x2n+1
arctan x = (−1)n (|x| ≤ 1).
n=0
2n + 1
2
Substitute x 7→ x : ∞
2
X x4n+2
arctan(x ) = (−1)n .
n=0
2n + 1
Thus R = 1.
(2) From the table,
∞
X x2n
cos x = (−1)n .
n=0
(2n)!

Page 10
 Substitute x 7→ 2x and multiply by x:
∞ ∞ 2n
X
n (2x)
X 22n 2n+1
x cos(2x) = x (−1) = (−1)n x .
n=0
(2n)! n=0
(2n)!

So R = ∞.
(3) Rewrite
−1/2
x2

x x x
√ = q = 1+ .
4 + x2 2 1+ x2 2 4
4
P∞ −1/2 x2
Using the binomial series (1 + u)−1/2 =


n=0 n
un for |u| < 1 with u = 4
,
∞    2 n X ∞  1  2n+1
x x X − 21 x −2 x
√ = = .
4 + x2 2 n=0 n 4 n=0
n 22n+1

Binomial-coefficient calculations:

− 21 − 12 − 1 · · · − 12 − n + 1 1 3 2n−1
 1






−2 ···
= = (−1)n 2 2 2
.
n n! n!
In particular,
 1
− 12
 1  1
− 12
   
−2 1 −2 3 −2 5 35
= 1, =− , = , =− , = .
0 1 2 2 8 3 16 4 128
So the first few nonzero terms are
x x x3 3x5 5x7 35x9
√ = − + − + − ··· .
4 + x2 2 16 256 2048 65536

x2
Here 4
< 1, so R = 2.

(4) Using the hint,

∞
!
1 1 X (2x)2n
sin2 x = (1 − cos(2x)) = 1− (−1)n .
2 2 n=0
(2n)!

Since the n = 0 term cancels, we get

∞
X 22n−1 2n
sin2 x = (−1)n−1 x .
n=1
(2n)!

So R = ∞.

Page 11
Problem 9
Use the definitions

ex − e−x ex + e−x
sinh x = cosh x =
2 2

and the Maclaurin series for ex to show that

∞ ∞
X x2n+1 X x2n
a) sinh x = b) cosh x =
n=0
(2n + 1)! n=0
(2n)!

Problem 10
Use the Maclaurin series for cos x to compute cos 5◦ correct to five decimal places.

Solution: Convert to radians:

5π π
5◦ = = .
180 36
π
Let x = 36 ≈ 0.0872664626.
Using the Maclaurin series
x 2 x4 x6
cos x = 1 − + − + ··· ,
2! 4! 6!
we compute the first three nonzero terms numerically:
x2 x4
− ≈ −0.0038077177, ≈ 0.0000024173.
2! 4!
Thus π x2 x4
cos ≈1− + ≈ 0.9961946996.
36 2! 4!
For an alternating series with decreasing terms, the error is at most the first
omitted term:
x6
|R3 | ≤ ≈ 5.11 × 10−10 ,
6!
so the approximation is correct to five decimal places. Therefore,

cos 5◦ ≈ 0.99619.

Page 12
Problem 11
1
a) Use the binominal series to expand √ .
1 − x2
b) Use part (a) to find the Maclaurin series for sin−1 x.

Solution:
(a) Use the binomial series
∞  
k
X k n
(1 + u) = u , |u| < 1.
n=0
n

Take k = − 12 and u = −x2 :

∞  1
1 2 −1/2
X −2
√ = (1 − x ) = (−x2 )n .
1−x 2
n=0
n

Now compute the coefficients using the binomial coefficient

 
k k(k − 1) · · · (k − n + 1)
= , n = 0, 1, 2, . . .
n n!

For k = − 21 ,
− 12 − 12 − 1 · · · − 12 − (n − 1)
 1



−2
= .
n n!
Multiplying by (−1)n makes all factors positive:
1
 1
3
5
2n−1


−2 · · · 1 · 3 · 5 · · · (2n − 1)
(−1)n = 2 2 2 2
= .
n n! 2n n!
Therefore,
∞  1
1 X −2 1 3 5
√ = (−x2 )n = 1 + x2 + x4 + x6 + · · · , (|x| < 1).
1 − x2 n=0
n 2 8 16

(b) Note that

d
1
arcsin x = √ .
dx 1 − x2
Integrate the series from part (a) term-by-term on [0, x] and use arcsin 0 = 0:
∞
xX ∞X 1 · 3 · 5 · · · (2n − 1) x2n+1
1 · 3 · 5 · · · (2n − 1) 2n
Z
arcsin x = t dt = · .
0 n=0
2n n! n=0
2n n! 2n + 1

That is,
x3 3x5 5x7
arcsin x = x + + + + ··· , (|x| < 1).
6 40 112

Page 13
Problem 12
Evaluate the indefinite integral as an infinite series.
Z √
cos x − 1
Z
1) 3
1 + x dx 2) dx
x

Solution:
(1) Use the binomial series
X∞ 1
1/2 2 un ,
(1 + u) = |u| < 1,
n=0
n

with u = x3 :
∞ 1
√ X 1 1 1 9 5 12
1+x = 3 2 x3n = 1 + x3 − x6 + x − x + ··· .
n=0
n 2 8 16 128
 
k k(k − 1)(k − 2) · · · (k − n + 1)
= , n = 0, 1, 2, . . .
n n!
 
k
= 1.
0
Coefficient calculations:
1 1
2 2
1
= 1, = ,
0 1 2
1 1 1 1
− 1 12 − 2
1
1



−1
 
2 2 2 1 2 2 2 1
= =− , = = ,
2 2! 8 3 3! 16
1 1
− 1 21 − 2 12 − 3
1



5
2 = 2 2 =− .
4 4! 128
Integrate term-by-term:
Z √ ∞  1  3n+1
X
2
x x4 x7 x10 5x13
1 + x3 dx = +C =x+ − + − + · · · + C.
n=0
n 3n + 1 8 56 160 1664

(2) Start from the Maclaurin series

∞
X x2n
cos x = (−1)n .
n=0
(2n)!
Then ∞
cos x − 1 X x2n−1 x x3 x5
= (−1)n =− + − + ··· .
x n=1
(2n)! 2! 4! 6!

Integrate term-by-term:
∞
cos x − 1 x2n x 2 x4 x6
Z X
dx = (−1)n +C =− + − + · · · + C.
x n=1
(2n)!(2n) 4 96 4320

Page 14
Problem 13
Use series to approximate the definite integral to within the indicated accuracy.
Z 1
2
1) x3 arctan x dx (four decimal places)
0
Z 0,4 √
2) 1 + x4 dx (|error| < 5 × 10−6 )
0

Solution:
P∞ n x2n+1
(1) Use arctan x = n=0 (−1) 2n+1 for |x| ≤ 1. Then
∞ 2n+4
n x
X
3
x arctan x = (−1) .
n=0
2n + 1

Let ∞
x2n+5
Z X
F (x) = x3 arctan x dx = (−1)n + C.
n=0
(2n + 1)(2n + 5)
By the Fundamental Theorem of Calculus,
Z 1/2 ∞
(1/2)2n+5
 
3 1 X
x arctan x dx = F − F (0) = (−1)n .
0 2 n=0
(2n + 1)(2n + 5)

This is an alternating series with decreasing terms, so the truncation error is at

most the first omitted term. Using the first few terms,
Z 1/2
(1/2)5 (1/2)7 (1/2)9 (1/2)11
x3 arctan x dx ≈ − + − .
0 (1)(5) (3)(7) (5)(9) (7)(11)
(1/2)13
The next term is < 10−5 , so this gives four-decimal-place accuracy.
(9)(13)
Numerically,
Z 1/2
x3 arctan x dx ≈ 0.0059.
0

(2) Use the binomial series

X∞ 1
1/2 2 un ,
(1 + u) = |u| < 1,
n=0
n

with u = x4 : ∞ 1
√ X
1 + x4 = 2 x4n .

n=0
n
Let Z √ ∞  1  4n+1
X
2
x
G(x) = 1+ x4 dx = + C.
n=0
n 4n + 1

Page 15
By the Fundamental Theorem of Calculus,
Z 0.4 √ ∞ 1
X
2
(0.4)4n+1
1 + x4 dx = G(0.4) − G(0) = .
0 n=0
n 4n + 1

This series is alternating for n ≥ 1 (since 1/2

n
alternates in sign) and has
decreasing terms for 0 ≤ x ≤ 0.4, so the error after truncating is at most the
first omitted term in absolute value.
Using terms up to n = 3:
Z 0.4 √
1 (0.4)5 1 (0.4)9 1 (0.4)13
1 + x4 dx ≈ (0.4) + − + .
0 2 5 8 9 16 13
The next term has magnitude
1
(0.4)17 5 (0.4)17
2 = < 5 × 10−6 ,
4 17 128 17

so the approximation satisfies |error| < 5 × 10−6 . Numerically,

Z 0.4 √
1 + x4 dx ≈ 0.400639.
0

Page 16
Problem 14
Use series to evaluate the limit.

x − ln (1 + x) 1 − cos x
1) lim 2) lim
x→0 x2 x→0 1 + x − ex

Solution:
x2 x3 4
(1) Use ln(1 + x) = x − − x4 + · · · for |x| < 1. Then
2
+ 3

x2 x3 x2 x3 x4
 
x − ln(1 + x) = x − x − + − ··· = − + − ··· ,
2 3 2 3 4
so
x − ln(1 + x) 1 x x2 1
2
= − + − · · · −−→ .
x 2 3 4 x→0 2
(2) Use
x2 x4 x2 x3
1 − cos x = − + ··· , ex = 1 + x + + + ··· .
2 24 2 6
Then
x2 x3 x2 x3
 
x
1+x−e =1+x− 1+x+ + + ··· =− − + ··· .
2 6 2 6
Therefore,
x 2 4 2
1 − cos x 2
− x24 + · · · 1
2
− x24 + · · ·
= = −−→ −1.
− 12 − x6 + · · · x→0
2 3
1 + x − ex − x2 − x6 + · · ·

Problem 15
Find the function represented by the given power series.
∞ 4n ∞
X
nx
X x2n+1
1) (−1) 2) (−1)n
n=0
n! n=0
22n+1 (2n + 1)

Solution:
(1) Recognize
∞ 4n ∞
X
nx
X (−1)n (x4 )n 4
(−1) = = e−x .
n=0
n! n=0
n!

(2) Let
∞ ∞ x 2n+1


X x2n+1 n
X
n 2
S(x) = (−1) 2n+1 = (−1) .
n=0
2 (2n + 1) n=0
2n + 1

Page 17
 P∞ n z 2n+1
Using arctan z = n=0 (−1) 2n+1 , we get
x
S(x) = arctan .
2

Problem 16
Find the sum of the series.
∞ ∞
X (−1)n X 3n
1) 2) (−1)n−1
n=0
n! n=1
n5n

Solution:
P∞ xn
(1) Using ex = n=0 n! , set x = −1:
∞
X (−1)n 1
= e−1 = .
n=0
n! e

(2) Rewrite the series:

∞ ∞ 3 n


n
X 3 X
(−1)n−1 n = (−1)n−1 5 .
n=1
n5 n=1
n
P∞ n−1 xn
Use ln(1 + x) = n=1 (−1) n
for |x| ≤ 1, x ̸= −1:
∞ 3 n

   
X 3 8
(−1)n−1 5
= ln 1 + = ln .
n=1
n 5 5

Page 18
Problem 17

Find the third-degree Taylor polynomial T3 (x) for the following functions centered
at the given points a, and graph both f (x) and T3 (x) on the same screen.
1. f (x) = ex , a=1
π
2. f (x) = cos x, a= 2

3. f (x) = ln x, a=1

Solution:

1. f (x) = ex , a=1
The Taylor polynomial for f (x) = ex centered at a = 1 is:
f ′′ (1) f ′′′ (1)
T3 (x) = f (1) + f ′ (1)(x − 1) + (x − 1)2 + (x − 1)3
2! 3!
Since f (x) = ex , all derivatives at x = 1 are equal to e:
e e
T3 (x) = e + e(x − 1) + (x − 1)2 + (x − 1)3
2 6
π
2. f (x) = cos x, a= 2

The Taylor polynomial for f (x) = cos x centered at a = π2 is:

′′ π ′′′ π



π  π  π f π f π
T3 (x) = f + f′ (x − ) + 2
(x − )2 + 2
(x − )3
2 2 2 2! 2 3! 2
The derivatives are:
π  π  π  π 
f = 0, f′ = −1, f ′′ = 0, f ′′′ =1
2 2 2 2
Thus, the Taylor polynomial is:

π (x − π2 )3
T3 (x) = −(x − ) +
2 6

Page 19
3. f (x) = ln x, a=1
The Taylor polynomial for f (x) = ln x centered at a = 1 is:
f ′′ (1) f ′′′ (1)
T3 (x) = f (1) + f ′ (1)(x − 1) + (x − 1)2 + (x − 1)3
2! 3!
The derivatives at x = 1 are:

f (1) = 0, f ′ (1) = 1, f ′′ (1) = −1, f ′′′ (1) = 2

Thus, the Taylor polynomial is:

(x − 1)2 (x − 1)3
T3 (x) = (x − 1) − +
2 3

Page 20
Figure 1: The graphs of each function along with their third-degree Taylor polynomials

Page 21
Problem 18
(a) Approximate f by a Taylor polynomial with degree n at the number a.
(b) Use Taylor’s Inequality to estimate the accuracy of the approximation f (x) ≈
Tn (x) when x lies in the given interval.
1) f (x) = 1/x, a = 1, n = 2, 0.7 ≤ x ≤ 1.3
2) f (x) = x2/3 , a = 1, n = 3, 0.8 ≤ x ≤ 1.2
2
3) f (x) = ex , a = 0, n = 3, 0 ≤ x ≤ 0.1

Solution:
(1) f (x) = x−1 , a = 1, n = 2.

′ f ′′ (1)
T2 (x) = f (1) + f (1)(x − 1) + (x − 1)2 = 1 − (x − 1) + (x − 1)2 .
2!
For R2 (x), f (3) (x) = −6x−4 , so on [0.7, 1.3],
6
M = max |f (3) (t)| = , |x − 1| ≤ 0.3,
0.74
hence
M 3 6/0.74 3 0.33
|R2 (x)| ≤ |x − 1| ≤ (0.3) = ≈ 0.11245.
3! 6 0.74
(2) f (x) = x2/3 , a = 1, n = 3.
2 1 4
T3 (x) = 1 + (x − 1) − (x − 1)2 + (x − 1)3 .
3 9 81
56 −10/3
Here f (4) (x) = − 81 x , so on [0.8, 1.2],
56
M = max |f (4) (t)| = 0.8−10/3 , |x − 1| ≤ 0.2,
81
and thus
M 56
0.8−10/3
|R3 (x)| ≤ |x − 1|4 ≤ 81
(0.2)4 ≈ 1.01 × 10−4 .
4! 24

Page 22
 2
(3) f (x) = ex , a = 0, n = 3.
T3 (x) = 1 + x2 .
2
We have f (4) (x) = (12 + 48x2 + 16x4 )ex , so on [0, 0.1],

M = f (4) (0.1) = 12.4816 e0.01 .

Therefore, for 0 ≤ x ≤ 0.1,

M 4 12.4816 e0.01
|R3 (x)| ≤ |x| ≤ (0.1)4 ≈ 5.25 × 10−5 .
4! 24

Problem 19
Use the Alternating Series Estimation Theorem or Taylor’s Inequality to estimate
the range of values of x for which the given approximation is accurate to within the
stated error. Check your answer graphically.

x3
sin x ≈ x − (|error| < 0.01)
6

Solution: Using the Maclaurin series

x3 x5 x7
sin x = x − + − + ··· ,
3! 5! 7!
3
the approximation x − x6 is obtained by truncating after the x3 term. For an alter-
nating series with decreasing term magnitudes, the error is at most the first omitted
term:
|x|5 |x|5
|error| ≤ = .
5! 120
|x|5
Require 120
< 0.01, so
√
|x|5 < 1.2
5
=⇒ |x| < 1.2 ≈ 1.037.

Therefore the approximation is accurate to within 0.01 for

√
5
√5
− 1.2 < x < 1.2 (approximately − 1.037 < x < 1.037).

(Alternative: Taylor’s Inequality.) For f (x) = sin x, the 4th derivative is f (4) (x) =
sin x, so |f (4) (x)| ≤ 1 for all x. Taylor’s Inequality for R3 gives

1 4 |x|4
|R3 (x)| ≤ |x| = .
4! 24
|x|4
Requiring < 0.01 gives
24 √
4
|x| < 0.24 ≈ 0.700.

Page 23
(Taylor’s Inequality for n = 4.) Since f (5) (x) = cos x and |f (5) (x)| ≤ 1, Taylor’s
Inequality gives
1 |x|5
|R4 (x)| ≤ |x|5 = .
5! 120
|x|5
Requiring < 0.01 yields
120 √
5
|x| < 1.2 ≈ 1.037.
Comparison.√ The alternating-series estimate and Taylor’s Inequality for n = 4 agree
on |x| < 5 1.2 ≈ 1.037, while Taylor’s Inequality for n = 3 is more conservative
3
(|x| < 0.700). Note that, T3 (x) = T4 (x) = x − x3! .

Page 24