2 Section 1.

2.  Compute each of the following limits and determine the corresponding rate of 
convergence.
ex −1
(a) limx→0 x
(b) limx→0 sinx x
x
(c) limx→0 e −cos x2
x−x

x2 /2−x4 /24
(d) limx→0 cos x−1+1+x
x6

(a)  From Taylor’s Theorem, ex = 1 +  x  + 21 x2 eξ for some ξ 

 x + ξ  between
 between 0 and x.
Hence,
ex 1  1 −
= 1 + xeξ .
x 2
Because
ex 1 
1 − − 
1 = x eξ < x || ||
x 2 
for all x
all  x  satisfying x <  ln 2, it follows that
 | |
ex − 1 = 1   with rate of convergence  O
lim  O((x).
x→0 x
3
(b)  From Taylor’s Theorem, sin x = x x6 cos ξ ξ    for some ξ 
ξ    between 0 and x.
−
Then,
sin x  x 2
=1 cos ξ  −
x 6
and
sin x 
1
1 = x2 cos ξ 

1 2
− x . | |≤
x 6 6
Finally,
sin x
lim  O((x2 ).
= 1   with rate of convergence  O
x→0 x

(c)  From Taylor’s Theorem, we have

 1  1
ex = 1 + x + x2 + x3 eξ 1

2 6
and
cos x  = 1 −  12 x2 +  16 x3 sin ξ 2
for some ξ 
some  ξ 1 and
and ξ 
 ξ 2  between 0 and x. Then
ex
− cos x − x = 1 +  x eξ − sin ξ 2 . 1

x2 6
For sufficiently small  x x,,  e ξ <  2
 2,, so |eξ − sin ξ 2 | <  2 + 1 = 3.
1 1
3. Thus,
 ex − cos x − x   |x|  ξ
 x2
− 1 = e − sin ξ 2 ≤ 1 |x|,
 6
1

2
 Convergence 3

and
ex − cos x − x = 1
lim with rate of convergence
convergence  O
 O((x).
x→0 x2
(d)  From Taylor’s Theorem, we have

cos x  = 1 −  12 x2 + 241 x4 − 720

1 6 1
x + x8 cos ξ 
8!
for some ξ 
some  ξ  between
 between 0 and  x
 x.. Hence,

cos x − 1 + 21 x2 − 241 x4 = − 1 1
 + x2 cos ξ,
x6 720 8!
and
1 2
 cos x − 1 + x − 1 4
24 x 1

1 2
x
2
6
+
720
=
8!x cos ξ | | ≤ 8!1 |x2|.
It therefore follows that

lim
cos x − 1 + 21 x2 − 241 x4 = − 1
x→0 x6 720
 O((x2 ).
with rate of convergence  O

3. Numerically determine which

which of the following sequences approaches
approaches 1 faster, and
then confirm the numerical evidence by determining the rate of convergence of 
each sequence.
sin x2 (sin x)2
lim v e rsus li m .
x→0 x2 x→0 x2

sin x2
The values in the following table suggest that x2 converges towa
toward
rd 1 more rapidly
2

than   (sinx2x) .

sin x2 (sin x)2

x
x2 x2
1.000
1.000 0.8414
0.8414709
709848
848079
0790
0 0.7080
0.7080734
734182
182735
7357
7
0.100
0.100 0.9999
0.9999833
833334
334166
1667
7 0.9966
0.9966711
711079
079379
3792
2
0.010
0.010 0.9999
0.9999999
999983
983333
3333
3 0.9999
0.9999666
666671
671111
1111
1
0.001
0.001 0.9999
0.9999999
999999
999998
9983
3 0.9999
0.9999996
996666
666667
6671
1
To confirm this conclusion, note that by Taylor’s Theorem,

sin u  =
 = u
 u −  16 u3 cos ξ,
for some ξ 
some  ξ  between
 between 0 and u
and  u.. Using the substitution  u  =  x 2 , we find
 = x

sin x2 =  x2 −  16 x6 cos ξ 

 4 Section 1.2

for some ξ   between 0 and  x 2 . Consequently

some  ξ  between Consequently,,
2
 sin x  1
x 2
− 1 = 6 x | cos ξ | ≤ 16 x .
4 4

from  f ((x) = (sin x)2 , we find

Starting from f 

f ′ (x) = 2sin x cos x  = sin2x,

sin2x, f ′′ (x) = 2cos
2cos 2x, f ′′′ (x) = −4sin2
4sin2x,
x,

 f (4) (x) =
and f 
and −8cos2
8cos2x
x. Therefore,

(sin x)2 = x 2 −  13 x4 cos2

cos2ξ 
ξ 

for some ξ 
some  ξ  between
 between 0 and  x
 x,, and
2
 (sin x)  1 2 1 2
 x 2
− 1 = 3 x | cos2
cos2ξ 
ξ | ≤ x .
3

Finally,
sin x2   (sin x)2
2
= 1 + O(x4 ) and 2
= 1 + O(x2 ).
x x

4.  Suppose that 0 <

0  < a < b.
b.
(a)  Show that if  α
 α n  =  α + O(1/nb ), then α
(1/n then  α n  =  α + O(1/na ).
(1/n
 f (x) =  L + O(xb ), then f 
(b)  Show that if  f ( then  f ((x) =  L + O(xa ).

(a)   Suppose αn = α +  + O O(1 /nb ). The

(1/n Then,n, there exists
exists a con
consta
stant
nt λ  such that for
1
sufficiently large  n
 n,, αn α  λ n . Beca
 | − | ≤ Because  a < b, it follows that  n a < nb
use a
b

and n1 > n1   for all n

a b all  n >  1
 1.. Thus,

|αn − α| ≤ λ n1b < λ n1a ,

and α
and  α n  =
 = α
 α + O(1/na ).
(1/n
Suppose  f ((x) =  L + O(xb ). Then, there exists a constant  K 
(b)   Suppose f   K  such
 such that for all
sufficiently small x
small  x,, f 
f ((x) L
 | K  x b . Because a
Because  a < b,
− |≤ | | b , it follows that for all
x 1, x b
| | ≤  | | ≤ | | x a . Thus, for sufficiently small  x
 x,,

|f f ((x) − L| ≤ K |x|b ≤ K |x|a,

 f ((x) =  L + O (xa ).
and f 
and

that  f 1 (x) =  L 1  + O(xa ) and f 

5.   Suppose that f  and  f 2 (x) =  L2  + O(xb ). Show that

c1 f 1 (x) + c2 f 2 (x) =  c 1 L1  + c2 L2  + O(xc ),

 Convergence 5

where c
where  c =
 = min(a,
min(a, b).

Suppose  f 1 (x) =  L1 + O(xa ) and

Suppose f   and f  =  L 2 + O(xb ). Then, there
 f 2 (x) = L there exist constants
constants
a
K 1  and
 and K 
 K 2  such that for all sufficiently small  x  x,, f 1 (x) L1
| − |≤ | | K 1 x and f 2 (x)
| −
b
L2 K 2 x . Let c
|≤ | | Let  c 1 and
and c
 c 2  be any real numbers. Applying the triangle inequality,
we find

|c1f 1(x) + c2f 2(x) − (c1L1 + c2L2)| ≤ |c1||f 1(x) − L1| + |c2||f 2(x) − L2|

≤ |c1|K 1|xa| + |c2|K 2|xb|.
Now, let c
let  c =
 = min(a,
min( a, b). Then, for  |x| <  1
 1,,

|c1|K 1|xa| + |c2|K 2|xb| < |c1|K 1|xc| + |c2|K 2|xc| = ( |c1|K 1 + |c2|K 2)|xc|.

Consequently,
c1 f 1 (x) + c2 f 2 (x) =  c 1 L1  + c2 L2  + O(xc ).

6.  The table below lists the errors of successive iterates for three different methods
√ 
3
for approximating 5. Estimate the order of convergence of each method, and
explain how you arrived at your conclusions.
Method 1 Method 2 Method 3
4.0 10−2
 × 3.7 10−4
 × 4.3 10−3
 ×
9.1 10−4
 × 1.2 10−15
 × 1.8 10−8
 ×
4.8 10−7
 × 1.5 10−60
 × 1.4 10−24
 ×
If a sequence converges of order α, then the error in each term of the sequence is
roughly the error in the previous term raised to the power α. Fr From
om the data
data for
for
“Method 1,” we see that each error is roughly the previous error squared; therefore,
we estimate the order of convergence to be α = 2. Fr From
om the data
data for “Method
“Method
2,” we see that each error is roughly the previous error raised to the fourth power;
therefore, we estimate the order of convergence to be  α
 α =
 = 4. Finally, from the data
for “Method 3,” we see that each error is roughly the previous error raised to the
third power; therefore, we estimate the order of convergence to be  α  = 3.

7. Let  pn  be a sequence which converges to the limit p.

 { }
(a) If 
lim
| p pn+1 − p| = 0,
n→∞ | p pn − p|α 0 ,
what can be said about the order of convergence of   pn  to p?  { }
(b) If 
lim
| p pn+1 − p| → ∞,
n→∞ | p pn − p|α
what can be said about the order of convergence of   pn  to p?  { }
 6 Section 1.2

(a) If 
lim
| p pn+1 − p| = 0,0 ,
n→∞ | p pn − p|α
then the numerator
numerator appapproach
roaches
es zero fast
faster
er than the denominato
denominator. r. In order
to achieve a nonzero limit, we must increase the power in the denominator.
Therefore
Therefore,, the order of convergence
convergence must be grea
greater
ter than  α
 α..
(b) If 
 pn+1  p
 p| − | →∞
lim ,
n→∞  p  pn  p α
| − |
then the denominator
denominator approaches
approaches zero faster than the numer
numerato
ator.
r. In order
to achieve a nonzero limit, we must decrease the power in the denominator.
Therefore, the order of convergence must be less than  α α..

8.  Suppose theory indicates that the sequence  pn   converges to p   of order 1.5.
 { }
Explain how you would numerically verify this order of convergence.

To numerically verify the order of convergence, calculate the ratio

| p pn+1 − p|
| p pn − p|1.5
for several successive values of  n. If the order
order of converge
convergence
nce is α = 1.5, these
ratios should approach a constant, specifically the asymptotic error constant.

9.  Theory indicates that the following sequence should converge to 3 of order

√ 
1.618.
1.618. Does the seque
sequence
nce actually
actually achieve
achieve an order of convergen
convergence
ce of 1.618
1.618?? If 
not, what is the actual order?
n  pn
0 2.0000
2.0000000
000000
000000
00000
00
1 1.6666
1.6666666
666666
666666
66667
67
2 1.7272
1.7272727
727272
272727
72727
27
3 1.7321
1.7321428
428571
571428
42857
57
4 1.7320
1.7320506
506804
804317
31722
22
5 1.7320
1.7320508
508075
075654
65499
99

Because the values in the third column of the following table appear to be approach-
ing a constant, the evidence suggests that the sequence does, in fact, converge
√ 
toward 3  with order of convergence  α α =
 = 1.618
618..
n pn | p pn − √ 3|/| p pn 1 − √ 3|1.618
−

1 2.0000
2.0000000
000000
000000
00000
00
2 1.666
1.66666
6666
6666
6666
6666
6667
67 0.5506
0.550660
6002
0295
9531
3142
42
3 1.727
1.72727
2727
2727
2727
2727
2727
27 0.39
0.3942
4292
9299
9985
8515
1516
16
4 1.732
1.73214
1428
2857
5714
1428
2857
57 0.52
0.5235
3588
8803
0316
1628
2884
84
5 1.732
1.73205
0506
0680
8043
4317
1722
22 0.43
0.4310
1007
0791
9144
4414
1420
20
6 1.732
1.73205
0508
0807
0756
5654
5499
99 0.48
0.4852
5255
5581
8157
5793
9327
27
 Convergence 7

10.  Theory indicates that the following sequence should converge to 4 /3 of order
1.618.. Does the seque
1.618 sequence
nce actually
actually achieve
achieve an order of convergen
convergence
ce of 1.61
1.618?
8? If 
not, what is the actual order?
n  pn
0 1.4986640
1.4986640985
985800
80016
16
1 1.4973
1.4973539
539977
977922
92205
05
2 1.4288
1.4288019
019773
773353
35339
39
3 1.4010
1.4010929
929153
153895
89552
52
4 1.3764
1.3764936
936760
760514
51456
56
5 1.3613
1.3613457
457455
455731
73130
30
6 1.3510
1.3510344
344825
825008
00881
81
7 1.3444
1.3444798
798506
506950
95066
66

Because the values in the third column of the following table are increasing with
n, the evidence suggests that the sequence does not have order of convergence
α = 1.618
618,, but rathe
ratherr that the order of convergence
convergence is less than 1.618. Becau
Because
se
the values in the fourth column appear to be approaching a constant, these values
suggest that the sequence is converging to  4
 4/
/3  with order of convergence  α
 α =
 = 1.

n pn | p pn − 4/3|/| p pn 1 − 4/3|1.618 | p pn − 4/3|/| p pn 1 − 4/3|

− −

1 1.4986
1.4986640
640985
985800
80022
2 1.49735399779221 3.01718763541581 0.99207588021590
3 1.42880197733534 1.77891367138598 0.58205253781266
4 1.40109291538955 3.03079120639280 0.70975745769255
5 1.37649367605146 3.36181849329742 0.63696294174768
6 1.36134574557313 4.52671513900300 0.64903127444432
7 1.35103448250088 5.75689539760301 0.63190377951100
8 1.34447985069507 7.61855893491390 0.62970586012393

11.  Show that the convergence of the sequence generated by the formula
x3n  + 3x
3 xn a
xn+1  = 2
3xn  + a
toward
 √ a  is third-order. What is the asymptotic error constant?
Note

xn+1
3
− √ a = xn + 3x
3 xn a √ 
− a =
− 3x2n√ a + 3x
x3n 3xn a − a3/2
3x2n  + a 3x2n  + a
√ 
(x − a)3
n
= .
3x2n  + a
Thus,
lim
|xn+1 −√ √ a|  = lilimm 1
=
1
.
n→∞ |xn − a|3 n 2
→∞ 3x  + a 4a
n
 8 Section 1.2

Consequently, xn  →  √ a  with order of convergence α = 3  and asymptotic error

 λ  = 41a .
constant λ
constant

12. Let a   be a non

non-ze
-zero
ro real number.
number. For any x0  satisfying 0 < x0 < 2/a
/a,, the
recursive sequence defined by

xn+1  =  xn (2 − axn)

converges to 1/a
1 /a.. What are the order of con
conver
vergenc
gencee and the asymptotic
asymptotic error
constant?

Note

xn+1 −  1a = xn(2 − axn) − a1 = −ax2n + 2x

2 xn −
1
a
   2

2 1  1
= −a x2n − xn  + 2  = −a xn − .
a a a

Thus,
lim
|xn+1 − a1 |  = lim a =
n→∞ |xn − a1 |2 n  = a.  a.
→∞

1
Consequently, xn → a   with order
order of convergence
convergence α = 2   and asymp
asymptoti
toticc erro
errorr
constant λ
constant  λ  =
 = a
 a..

13.  Suppose that the sequence  pn conv

converges
{ }erges linearly to the limit  p  with asymptotic
error constant λ
constant  λ.. Further suppose that  p n+1  p
 p,,  p n  p
 p and
 and  p n−1  p
− −  p are
 are all of 
−
the same sign. Show that
 pn+1  pn − ≈
λ.
 pn  pn−1 −
Suppose the sequen
sequence
ce  pn  converges linearly to  p  with asymptotic error constant
 { }
λ. Then
lim
 pn+1  p
 p |
=  λ,
− |
n→∞  p  pn  p | − |
so, for sufficiently large  n
 n,,

| p pn+1 − p| ≈ λ| p pn − p|.

Moreover,
| p pn − p| ≈ λ| p pn 1 − p| or | p pn 1 − p| ≈ λ1 | p pn − p|.
− −

Because we are given that  p n+1 − p  p,,  p n − p

 p and
 and  p n 1 − p
 p are
 are all of the same sign,
−

we may drop the absolute values

values from the above exp
expressio
ressions.
ns. Now
Now,,

 pn+1  pn
− pn+1  p ( pn
− − − p p))
=
 pn  pn−1
−  pn  p ( pn−1
− − − p p))
 Convergence 9

λ( pn  p
 p)) ( pn  p
− − −  p))
≈ 1
 pn  p λ ( pn  p)
− − −  p)
λ 1
−
= =  λ.
1 λ1
−

14.   A sequence  pn  converges   superlinearly  to

 { }  to p  provided

lim
| p pn+1 − p| = 0.0 .
n →∞ | p pn − p|
Show that if  p
 p n →  p  of order α
 p of order  α  for  α >  1, then  { pn }  converges superlinearly to
p.

Suppose the sequence  pn   converges p   of order α > 1   with asymp

 { } asymptoti
toticc erro
errorr
constant λ
constant  λ.. Then
lim
 pn+1  p
 p | − |
=  λ.
n→∞  p pn  p α | − |
Then

lim
| p pn+1 − p|   = lim
| p pn+1 − p| · | p pn − p|α 1
−

n→∞ | p pn − p| n→∞ | p pn − p|α

= lim
| p pn+1 − p| · lim | p pn − p|α
− 1
n→∞ | p pn − p|α n
→∞

= λ 0 = 0.·
Therefore,  pn  converges superlinearly to p
 { } to  p..

15.  Suppose that { pn }  converges superlinearly to p  (see Exercise 14). Show that

lim
| p pn+1 − pn| = 1.1 .
n →∞ | p pn − p|
Note that
 pn+1  pn
− pn+1  p − − ( pn − p p)) = pn+1 − p − 1.
=
 pn  p
−  pn − p  pn − p
Because  pn  converges superlinearly to p
 { } to  p,, it then follows that

lim
| p pn+1 − pn| =  lim  pn+1 − p − 1 = |0 − 1| = 1.1 .
n→∞ | p pn − p| n →∞  pn − p 
(a))  Determine the third-degree Taylor polynomial and associated remainder
16.. (a
16
term for the function f 
function  f ((x) = ln(1 x). Use x
Use  x 0  = 0.
−
10 Section 1.2

(b)  Using the results of part (a), approximate ln(0 .25) and compute the the-
oretical
oretical error bound associated with this approximat
approximation.
ion. Comp
Compare
are the
theoretic
theor etical
al error bound with the actua
actuall error
error..
(c)  Compute the following limit and determine the corresponding rate of con-
vergence:
ln(1 x) + x + 21 x2 −
lim .
x→0 x3

(a) Let
Let f 
 f ((x) = ln(1
− x). Then
1 1
′
f  (x) = −
1−x
′′
, f  (x) = −
(1 − x)2
, f  ′′′
(x) = − (1 −2 x)3 , and f (4)(x) = − (1 −6 x)4 .
Moreover,

f (0)
f (0) = ln 0 , f ′ (0) =
ln 1 = 0, ′′
−1, f  (0) = −1, f  ′′′
(0) = −2, and f (4)(ξ ) = − (1 −6 ξ )4 .
Finally,
ln(1 − x) = P 3 (x) + R3 (x)
 x 2  x 3 4
= x − − − − 4(1 x− ξ )4 ,
2 3
for some ξ 
some  ξ  between
 between 0 and x
and  x..
(b)  Using the result of part (a),
 0..752
 0  0..753
 0
ln(0..25)
ln(0 ≈ P 3(0(0..75) = −0.75 − − −1.171875
171875.. =
2 3
Because 0
Because  0  < ξ <  0 75,,  (1 − ξ ) 4 ≤ 44 and
 0..75 −

0.754 81
|error| = |R3(0 (0..75)| =
4(1 − ξ )4
 ≤ 4
= 20.
20 .25
25..

The actual error is  | ln(0

ln(0..25) − P 3 (0(0..75)| ≈ 0.214419
214419,, which is significantly
less than the theoretical
theoretical error bound.
(c)  Once again using the result from part (a), we find
ln(1 − x) + x + 21 x2 = − 1 − x
x3 3 4(1 − ξ )4 .
Moreover,  ln(1 − x) + x + 1 2
2x

 1 x | |  ≤ |x|,
3
+ =
 x 3  4 1 ξ  4 | − |
for all sufficiently small  x
 x.. Therefore,

lim
ln(1 − x) + x + 21 x2 = − 1 ,
x→0 x3 3
with rate of convergence  O
 O((x).
 Convergence 11

17 (a))  Determine the third-degree Taylor polynomial and associated remainder

17.. (a √ 
term for the function f 
function  f ((x) = 1 + x. Use x
Use  x 0  = 0.
(b) Using the results of part (a), approximate
√ 
approximate 1.5 and compute the theoretical
error bound associated with this approximation. Compare the theoretical
error bound with the actual error.
(c)  Compute the following limit and determine the corresponding rate of con-
vergence: √ 
1 + x 1 21 x − −
lim .
x→0 x2

(a) Let
Let f 
 f ((x) =
√ 1 + x. Then
1
f ′ (x) =
2
(1 + x)−1/2 , f ′′ (x) = − 14 (1 + x) − 3/2
, f ′′′(x) = − 38 (1 + x) 5/2
−
,

 f (4) (x) =
and f 
and − 1516 (1 + x) − 7/2
. Moreover,
1 ′′
f (0)
f  1, f ′ (0) =
(0) = 1,
2
, f  (0) = − 14 , f  ′′′
(0) =
3
8
,

 f (4) (ξ ) =
and f 
and − 1516 (1 + ξ ) −7/2
. Finally
Finally,,
√ 1 + x = P 3 (x) + R3 (x)
 1  1 2 1 5 4
= 1+ x x + x3 − − 128 x (1 + ξ ) − 7/2
,
2 8 16
for some ξ 
some  ξ  between
 between 0 and  x
 x..
(b)  Using the result of part (a),
√   1  1 1
1.5
≈ P 3(0
(0..5) = 1 + (0 (0..5) − (0 (0..5)2 + (0 (0..5)3 = 1.
1 .2265625
2265625..
2 8 16
Because 0
Because  0 <  0..5,  (1 + ξ ) 7/2 ≤ 1  and
 < ξ <  0 −

5
|error| = |R3(0 (0..5)| ≤ (0..5)4 ≈ 2.44 × 10 3 .
(0 −

128
√ 
The actual error is  | 1.5 − P 3 (0 (0..5)| ≈ 1.82 × 10 3 , which is less than the
−

theoretical error bound.

(c)  Once again using the result from part (a), we find
√ 1 + x − 1 − 1 x
x2
2
= − 18 − 16x (1 + ξ ) −5/2
.

Moreover,
 √ 1 + x − 1 − 1

2x  1
 x 2
+
8
=
x
16
1
  | | | − ξ | − 5/2
≤ 161 |x|,
 12 Section 1.2

for all sufficiently small  x

 x.. Therefore,
√ 1 + x − 1 − 1 x
x→0
lim
x2
2
= − 18 ,
with rate of convergence  O
 O((x).

In Exercises 18 - 21, verify that Taylor’s

Taylor’s theorem produces the indicated formula,
formula,
where ξ 
where  ξ  is
 is between 0 and  x .

18.
 x 2 n n+1
ex = 1 + x +
2
+ · · · +  xn! + (nx+ 1)! eξ

Let f  Then  f (n) (x) =  e x and

Let f ((x) =  ex . Then f   f (n) (0) = 1 for
and f  1  for all n
all  n.. Therefore, by Taylor’s
Theorem with x
with  x 0  = 0,
n (k)
x
 f  (0) k  f (n+1) (ξ ) n+1
e = x + x
k! (n + 1)!
k=0
 x 2 n n+1
= 1+x+
2
+ · · · +  xn! + (nx+ 1)! eξ ,
for some ξ 
some  ξ  between
 between 0 and  x
 x..

19.
3 5 n 2n+1
 ( 1)n+1 x2n+3
sin x  =
 = x
 x −  x3! +  x5!  − + · · · +  ( −(2
1) x
(2nn + 1)!
+
−
(2n
(2n + 3)!
cos ξ 

Let f 
Let  f ((x) = sin x. Then
f ′ (x) = cos x, f ′′ (x) = − sin x, and
and f 
 f  ′′′
(x) = − cos x.
Moreover,
f (0) 0 , f ′ (0) = 1,
f (0) = 0, 1, f ′′ (0) = 0,
0, and  f ′′′(0) =
and f  −1.
As higher-order derivatives are calculated, this cycle of four values repeats indefi-
nitely. In particular, we find
f (2
(2n
n)
(0) = 0 and f (2
(2n
n+1)
(0) = ( 1)n . −
Therefore, by Taylor’s Theorem with  x 0  = 0,
3 5 n 2n+1
 x  x  ( −1) x  ( 1)n+1 x2n+3
−
sin x  =
 = x
 x − +  − + · · · + + cos ξ,
3! 5! (2n
(2n + 1)! (2n
(2n + 3)!
for some ξ 
some  ξ  between
 between 0 and  x
 x..
 Convergence 13

20.
2 4 n 2n
 ( 1)n+1 x2n+2
cos x  = 1 −  x2! +  x4! − + · · · +  ( −(2
1) x
(2n
n)!
+
−
(2n
(2n + 2)!
cos ξ 

Let f 
Let  f ((x) = cos x. Then

f ′ (x) = ′′
− sin x, f  (x) = − cos x, and
and f 
 f  ′′′
(x) = sin x.

Moreover,
f (0)
f  1, f ′ (0) = 0,
(0) = 1, 0, f ′′ (0) = −1, and
and f 
 f  ′′′
(0) = 0.
0.
As higher-order derivatives are calculated, this cycle of four values repeats indefi-
nitely.. In particular, we find
nitely

f (2
(2nn)
(0) = ( 1)n and f (2
− (2n
n+1)
(0) = 0.
0.

Therefore, by Taylor’s Theorem with  x 0  = 0,

2 4 n 2n
 ( 1)n+1 x2n+2
cos x  = 1 −  x2! +  x4!  − + · · · +  ( −(2
1) x
(2nn)!
+
−
(2n
(2n + 2)!
cos ξ,

for some ξ 
some  ξ  between
 between 0 and x
and  x..

21.
n+1 n+1
1
1+x
=1 − x + x2 − + · · · + (−1)nxn +  ( −(11)+ ξ )nx+2

1
Let f 
Let  f ((x) = 1+x  =
1+x (1 + x)−1 . Then,
2
f ′ (x) = −1 · (1 + x) −
, f ′′ (x) = 1 2 (1 + x)−3 , f ′′′(x) =
· · −
−1 · 2 · 3 · (1 + x) 4
,

general,  f (n) (x) = (−1)n · n! · (1 + x)

and, in general, f  − n−1
. Therefore, by Taylor’s Theorem
with x
with  x 0  = 0,
n+1 n+1
1
1+x
=1 − x + x2 − + · · · + (−1)nxn +  ( −(11)+ ξ )nx+2 ,

for some ξ 
some  ξ  between
 between 0 and x
and  x..