Sequences and Limit of Sequences: 4.1 Sequences: Basic de Nitions

Chapter 4
Sequences and Limit of

Sequences
4.1 Sequences: Basic Denitions
Denition 278 (sequence) A sequence is a function whose domain is a subset
of the form
: Z : : _ :
0
for some :
0
Z
The elements or the terms of the sequence, usually denoted r
n
will be of the
form: r
n
= ) (:). If the terms of a sequence are denoted r
n
, then the sequence
is denoted (r
n
) or r
n
. We can think of a sequence as a list of numbers. In this
case, a sequence will look like: r
n
= r
1
, r
2
, r
3
, .... The starting point, :
0
is
usually 1 but it does not have to be. However, it is understood that whatever
the starting point is, the elements r
n
should be dened for any : _ :
0
. For
example, if the general term of a sequence is r
n
=
2:
: 4
, then, we must have
:
0
_ 5.
A sequence can be given dierent ways.
1. List the elements. For example,
_
1
2
,
2
3
,
3
4
, ...
_
. From the elements listed,
the pattern should be clear.
2. Give a formula to generate the terms. For example, r
n
= (1)
n
2
n
:!
. If
the starting point is not specied, we use the smallest value of : which
will work.
3. A sequence can be given recursively. The starting value of the sequence
is given. Then, a formula to generate the :
th
term from one or more
103
104 CHAPTER 4. SEQUENCES AND LIMIT OF SEQUENCES
previous terms. For example, we could dene a sequence by giving:
r
1
= 2
r
n+1
=
1
2
(r
n
+ 6)
Another example is the Fibonacci sequence dened by:
r
1
= 1, r
2
= 1
r
n
= r
n1
+ r
n2
for : _ 3
Like a function, a sequence can be plotted. However, since the domain is a
subset of Z, the plot will consist of dots instead of a continuous curve.
Since a sequence is dened as a function. everything we dened for functions
(bounds, supremum, inmum, ...) also applies to sequences. We restate those
denitions for convenience.
Denition 279 (Bounded Sequence) As sequence (r
n
) is said to be bounded
above if its range is bounded above. It is bounded below if its range is bounded be-
low. It is bounded if its range is bounded. If the domain of (r
n
) is : Z : : _ / for some integer /
then the above denition simply state that the set r
n
: : _ / must be bounded
above, below or both.
Denition 280 (One-to-one Sequences) A sequence (r
n
) is said to be one-
to-one if whenever : ,= : then r
n
,= r
m
.
Denition 281 (Monotone Sequences) Let (r
n
) be a sequence.
1. (r
n
) is said to be increasing if r
n
_ r
n+1
for every : in the domain of the
sequence. If we have r
n
< r
n+1
, we say the sequence is strictly increasing.
2. (r
n
) is said to be decreasing if r
n
_ r
n+1
for every : in the domain of
the sequence.
3. A sequence that is either increasing or decreasing is said to be monotone.
If it is either strictly increasing or strictly decreasing, we say it is strictly
monotone.
Denition 282 (constant) (r
n
) is constant if r
n+1
= r
n
\:. (r
n
) is eventu-
ally constant if r
n+1
= r
n
\: for some positive integer .
Denition 283 (Periodic) (r
n
) is periodic if j Z
+
such that \:, r
n+p
=
r
n
.
Sometimes, when studying at a sequence, it is useful to look only at some
of its terms, not all of them. For example, we could take every other term. We
could also take some of the terms at random. More specically, given a sequence
a
n
= a
1
, a
2
, a
3
, ..., we can dene a new sequence /
k
= a
n
k
, made from
4.2. LIMIT OF A SEQUENCE: DEFINITIONS 105
some of the terms of a
n
as follows: Let :
1
denote the index of the rst element
of a
n
we select and dene /
1
= a
n1
.. Let :
2
denote the index of the second
element of a
n
2
= a
n2
. In general, let :
k
denotes the
index of the /
th
element of a
n
k
= a
n
k
. In doing so,
note that :
k
_ / (as long as : _ 1 which we can achieve by renumbering).
Denition 284 (subsequence) The sequence /
k
as dened above is called a
subsequence of a
n
. Simply speaking, a subsequence is a sequence which only
contains some of the terms of the original sequence. However, it still contains
an innite amount of terms, in other words, it does not stop.
Another way of dening a subsequence is given below.
Denition 285 (subsequence) A subsequence of a sequence (r
n
) is a se-
quence of the form
_
r
'(n)
_
where , : N N is a strictly increasing function.
For example, (r
2n
) is a subsequence of (r
n
), in this case, ,(:) = 2:. (r
2n+1
)
is another subsequence, in this case ,(:) = 2: + 1.
Example 286 Given a sequence (r
n
)
1. (r
2n
) is a subsequence of (r
n
), in this case, ,(:) = 2:.
2. (r
2n+1
) is another subsequence, in this case ,(:) = 2: + 1.
4.2 Limit of a Sequence: Denitions
4.2.1 The Concepts of Eventually and Frequently
Denition 287 (Eventually and frequently) Let (r
n
) be a sequence and
o _ R.
1. (r
n
) is eventually in o if there exists 0 such that : _ ==r
n
o.
2. (r
n
) is frequently in o if r
n
o for innitely many values of :.
Examples:
(r
n
) where r
n
= (1)
n
is frequently in the set 1 and well as frequently
in the set 1.
(r
n
) where r
n
=
_
1 if 1 _ : _ 20
0 if : 20
is eventually in the set 0.
(r
n
) where r
n
= ln: is eventually is the set [2, ).
The following are simple observations regarding these two notions:
If a sequence is in a set, then it is eventually in the set.
If a sequence is eventually in a set, then it is frequently in the set.
(r
n
) is eventually in o if and only if it is not frequently in R o.
(r
n
) is frequently in o if and only if it is not eventually in R o.
The following is an important theorem regarding subsequences.
Theorem 288 Let (r
n
) be a sequence and o _ R. The following are equivalent:
1. (r
n
) is frequently in o.
2. (r
n
) has a subsequence in o.
Proof. We prove both directions.
1. (1 ==2). Since (r
n
) is frequently in o, (r
n
) o for innitely many :.
Let :
i
be the i
th
index of (r
n
) for which (r
n
) o. Then the sequence (j
n
)
where j
i
= r
ni
is both a subsequence of (r
n
) and is in o.
2. (2 ==1). Since a subsequence of (r
n
) contains an innite number of
terms of (r
n
), the result follows.
4.2.2 Denitions of Limits
In the following, we assume that (r
n
) is a sequence of real numbers and 1 is an
extended real number. In this section, We wish to investigate the behavior of
r
n
as : gets arbitrarily large, in other words as : . This is called nding
the limit of r
n
as : approaches , it is denoted lim
n!1
r
n
. We want to know if
the values of r
n
keep on changing as : , or if they seem to get closer to
some number. Several things can happen. As : , r
n
could get arbitrarily
large (or small), in other words, r
n
. Another possibility is that the
values of r
n
will get closer and closer to some number we will call 1. But, it is
also possible that the values of r
n
will keep on changing forever. If we allow 1
to be an extended real number, then the rst two cases can be summarized in
lim
n!1
r
n
= 1. We now give a more precise denition of limits as well as partial
limits.
Denition 289 (Limit) lim
n!1
r
n
= 1 if for every neighborhood l of 1, (r
n
)
is eventually in l. In this case, we write one of the following:
lim
n!1
r
n
= 1
limr
n
= 1
r
n
1 as :
r
n
1
If 1 is a real number, we say that (r
n
) converges to 1 or that 1 is a nite
limit for (r
n
). Otherwise, we sat that (r
n
) diverges. Note that (r
n
) may diverge
for two reasons. The rst one is if 1 = , in this case, we say that (r
n
) has
an innite limit. The second is if (r
n
) has no limit at all.
Denition 290 (Partial limit) 1 is a partial limit of (r
n
) if for every neigh-
borhood l of 1, (r
n
) is frequently in l.
We now give equivalent conditions for a sequence to have a limit and a
partial limit. In each case, we consider whether the limit (partial limit) is nite
or innite.
Theorem 291 (Finite Limit) Let (r
n
) be a sequence, and 1 R. The fol-
lowing are equivalent:
1. 1 is a limit of (r
n
) in other words lim
n!1
r
n
= 1.
2. \c 0, (r
n
) is eventually in (1 c, 1 + c).
3. \c 0, 0 such that : _ ==r
n
(1 c, 1 + c).
4. \c 0, 0 such that : _ ==[r
n
1[ < c.
Proof. We prove (1 ==2 ==3 ==4 ==1).
1. (1 ==2). Let c 0 be given. Then (1 c, 1 + c) is a neighborhood of
1. Thus, if lim
n!1
r
n
= 1 it follows by denition that (r
n
) is eventually in
(1 c, 1 + c).
2. (2 ==3). This is the denition of eventually.
3. (3 ==4). r
n
(1 c, 1 + c) ==1c < r
n
< 1+c ==c < r
n
1 <
c ==[r
n
1[ < c.
4. (4 ==1). We need to show that (r
n
) is eventually in l where l is any
neighborhood of 1. If l is any neighborhood of 1, then by denition, l
contains (1 c, 1 + c) for some c 0. We can nd 0 such that
: _ ==[r
n
1[ < c
== c < r
n
1 < c
== 1 c < r
n
< 1 + c
== r
n
(1 c, 1 + c)
== r
n
l
Thus r
n
is eventually in l. Since l was arbitrary, the result follows.
Remark 292 Part 4 of theorem 291 is known as the c c denition of the
limit.
Theorem 293 (Finite Partial Limit) Let (r
n
) be a sequence, and 1 R.
The following are equivalent:
1. 1 is a partial limit of (r
n
).
2. \c 0, (r
n
) is frequently in (1 c, 1 + c).
3. \c 0, r
n
(1 c, 1 + c) for innitely many :.
4. \c 0,[r
n
1[ < c for innitely many :.
Proof. See problems
Theorem 294 (Innite Limit) Let (r
n
) be a sequence, and 1 = (the case
1 = is similar and left to the reader). The following are equivalent:
1. is a limit of (r
n
) in other words lim
n!1
r
n
=
2. \' 0, (r
n
) is eventually in (', ).
3. \' 0, 0 such that : _ ==r
n
(', ).
4. \' 0, 0 such that : _ ==r
n
'
Proof. See problems
Theorem 295 (Innite Partial Limit) Let (r
n
) be a sequence, and 1 =
(the case 1 = is similar and left to the reader). The following are equiva-
lent:
1. is a partial limit of (r
n
).
2. \' 0, (r
n
) is frequently in (', ).
3. \' 0, r
n
(', ) for innitely many :.
4. \' 0, r
n
' for innitely many :.
5. (r
n
) is unbounded above.
Proof. See problems
4.2.3 Examples
Before we look at examples and elementary theorems, let us make some remarks.
[r
n
1[ simply represents the distance between r
n
and 1. If we think of c
as a very small quantity, then the above denition says that we can make
the distance between r
n
and 1 as small as we want, simply by taking :
large enough . In fact, no matter how small c is, we can make the distance
smaller. In essence, this is saying that if : is large enough, r
n
and 1
become identical. Think of this as a game between two persons. The
other person gives you a value for c, that it the other person tells you how
close he or she wants r
n
to be from 1. You must be able to nd such
that if : _ , r
n
will be within the prescribed distance. Not only r
n
, but
also all the terms after r
n
will be within the prescribed distance.
Graphically, if we draw the horizontal lines j = 1+c and j = 1c, then
saying : _ = [r
n
1[ < c means that all the values of r
n
will fall
within the region between the two lines whenever : _ .
It should be clear to the reader that to prove the limit is 1, c is arbitrary.
Given any c, the reader must prove the number in the denition can
be found. Usually, will depend on c. We now look at some examples to
see how this is done.
Example 296 Let (r
n
) be such that r
n
= c, for some constant c. Show that
r
n
c.
We need to show that for any c 0, (r
n
) is eventually in (c c, c + c). Since
r
n
= c for any :, (r
n
) is always in c _ (c c, c + c) for any c 0. Hence,
(r
n
) is eventually in (c c, c + c).
Example 297 This example is similar to the previous one, the dierence is
that (r
n
) is eventually constant, that is there exists 0 such that r
n
= c
whenever : _ . Again, show that r
n
c.
The proof is similar. Here, we have that (r
n
) is eventually in c _ (c c, c + c).
Example 298 Let (r
n
) be such that r
n
= (1)
n
. Prove that (r
n
) does not
have a limit (nite or innite) but has two partial limits: 1 and 1.
First, we prove that is not a limit of (r
n
). We can nd ' 0 such
that (r
n
) is not eventually in (', ). Any number ' 1 will do.
Next, we prove that is not a limit of (r
n
).We can nd ' < 0 such
that (r
n
) is not eventually in (, '). Any number ' < 1 will do.
Next, we prove that no real number can be a limit of (r
n
). We note that
r
n
only has two values, 1 and 1. For a real number r to be a limit of
(r
n
), both 1 and 1 would have to be eventually in (r c, r + c), for any
c 0. Clearly, if c =
1
2
, this cannot happen, no matter what r is. So, no
real number can be a limit of (r
n
).
Next, we prove that 1 is a partial limit. We need to prove that (r
n
) is
frequently in (1 c, 1 + c), for any c 0. By theorem 288, it is enough
to show (r
n
) has a subsequence in (1 c, 1 + c), for any c 0. The
subsequence (r
2n
) is in 1 _ (1 c, 1 + c), for any c 0. The result
follows.
Finally, we prove that 1 is a partial limit. This is similar. We simply
note that the subsequence (r
2n1
) is in 1.
Example 299 Let (r
n
) be such that r
n
= :. Prove that r
n
. We need to
show that for any ' 0, there exists 0 such that : _ ==r
n
'. We
want r
n
'. This will happen when : ' (since r
n
= :). So, we see that
if is any integer greater than ', we will have : ' == r
n
'. Hence,
the result.
Example 300 Let (r
n
) be such that r
n
= :
2
. Prove that r
n
. We need
to show that for any ' 0, there exists 0 such that : _ ==r
n
'.
We want r
n
'. This will happen when :
2
' (since r
n
= :
2
), that is
when :
_
'. So, we see that if is any integer greater than
_
', we will
have : ' ==r
n
'. Hence, the result.
Example 301 Let (r
n
) be such that r
n
= (1)
n
:
2
. Prove that (r
n
) has no
limit and that are partial limits.
1. First, we see why neither nor can be limits of (r
n
). (r
n
) cannot
be eventually in (', ) for any number ' 0 because if r
n
(', ),
r
n+1
, (', ). For the same reason, (r
n
) cannot be eventually in
(, ') for any number ' < 0.
2. Then, we see why a real number r cannot be a limit for (r
n
). Let r R.
For r to be a limit of this sequence, (1)
n
:
2
has to be eventually in
(r c, r + c) for any c 0. We consider several cases.
r = 0. [r
n
[ increases without bounds, so r
n
will eventually get out of
any interval of the form (c, c).
r ,= 0. Let c =
[r[
2
. If r
n
is in (r c, r + c) then r
n+1
cannot be
there (why?)
3. Finally, we see why and are partial limits of (r
n
). The subsequence
(r
2n
) is eventually (', ) for any number ' 0 thus (r
n
) is frequently
in (', ) for any number ' 0. This means that is a partial limit
of (r
n
). Similarly, the subsequence (r
2n+1
) is eventually in (, ') for
any number ' < 0. This means that is a partial limit of (r
n
).
Example 302 Show that the sequence whose general term is r
n
=
1
:
converges
to 0.
Given c 0, we need to nd such that : _ =
1
:
0
< c. To do this,
we usually start with what we want, i.e.
1
:
0
< c, and we try to nd the

condition : has to satisfy for it to happen.
1
:
0
< c =
1
:
< c
=
1
:
< c
=:
1
c
So, we see that if =
1
c
, then : _ =
1
:
0
< c.
Example 303 Show that the sequence whose general term is r
n
=
1
:
2
converges
to 0.
Given c 0, we need to nd such that : _ =
1
:
2
0
< c. To do this,
we usually start with what we want, i.e.
1
:
2
0
< c, and we try to nd the

condition : has to satisfy for it to happen.
1
:
2
0
< c =
1
:
2
< c
=
1
:
2
< c
=:
2
1
c
=:
1
_
c
or : <
1
_
c
Since : is a natural number, we only retain :
1
_
c
. So, we see that if =
1
_
c
,
then : _ =
1
:
0
< c.
Example 304 Same example as above, using a slightly dierent approach.
When we had
1
:
2
< c. We could have noticed that if : 1 we have
1
:
2
<
1
:
.
So, if we make
1
:
< c then what we need,
1
:
2
< c will also happen. We can
make
1
:
< c simply by taking :
1
c
, it will follow that
1
:
2
0
< c. This time,

we nd that =
1
c
works.
Remark 305 It seems that if we do the same problem dierent ways, we nd a
dierent solution. We really dont. When we nd , the only thing we guarantee
is that if : _ , then [r
n
1[ < c. It might also be true if : is slightly less
than . In other words, we are not nding the smallest that will work. If we
use approximations like in example 304 , we may not get the best estimate for
.
Remark 306 The idea is the following. Forget mathematics for a second (just
a second). Imagine you are driving on a dirt road and you come to a bridge.
You are not sure if the bridge can withstand the weight of your car. If ahead of
you is a vehicle you know to be heavier than yours which just cleared the bridge,
then you know your vehicle can clear the bridge as well.
Remark 307 The above example illustrates a general technique to nd given
c. In general, the general term of the sequence is too complicated for us to be
able to solve for :. If we call r
n
the general term of the sequence, we do the
following: Start with [r
n
1[ and simplify it as much as possible. We get an
expression involving :, call it 1
1
(:). If we still cannot solve for : in 1
1
(:) < c,
we try to nd an upper bounds of [r
n
1[, call it 1
2
(:) for which we can solve
1
2
(:) < c. The result will follow. Since [r
n
1[ = 1
1
(:) < 1
2
(:), if we can
have 1
2
(:) < c, we will also have [r
n
1[ < c. Obviously, 1
2
(:) has to be
small enough so we can make it as small as we want. We illustrate this with
the next example.
Example 308 Show that the sequence r
n
whose general term is r
n
=
2: 1
3: + 2
converges to
2
3
.
Again, as before, we look at what we need, and see if we can derive the condition
: must satisfy.
2: 1
3: + 2

2
3
< c ==
6: 3 6: 7
3 (3: + 2)
==
10
3 (3: + 2)
< c
Though here we could solve for :, we can also use an upper bound instead. We
notice that
10
3 (3: + 2)
=
10
9: + 6
<
10
9:
Thus, if we make
10
9:
< c, it will follow that
2: 1
3: + 2

2
3
< c. The former

happens when :
10
9c
. So, given c 0, :
10
9c
==
2: 1
3: + 2

2
3
< c this
proves that lim
2: 1
3: + 2
=
2
3
.
Example 309 Prove that lim
__
: + 1
_
:
_
= 0.
Again, as before, we look at what we need, and see if we can derive the condition
: must satisfy.
__
: + 1
_
:
_
0
< c ==
__
: + 1
_
:
_ __
: + 1 +
_
:
_
_
: + 1 +
_
:
< c
We work on the left side of the inequality and nd an upper bound.
__
: + 1
_
:
_ __
: + 1 +
_
:
_
_
: + 1 +
_
:
=
1
_
: + 1 +
_
:
<
1
2
_
:
So, if we make
1
2
_
:
< c, we will have what we need, that is

_
: + 1
_
:
< c.
The former happens when
_
:
1
2c
or :
1
4c
2
(since : 0). Now, we see
that given c 0, if :
1
4c
2
then

__
: + 1
_
:
_
0
< c thus proving that

lim
__
: + 1
_
:
_
= 0.
4.2.4 Exercises
1. Let r
n
= 3 +
1
:
. Prove that r
n
3.
2. Let r
n
= 3 +
2
:
. Prove that r
n
3.
3. Let r
n
=
1
:
for : N. If r R : r ,= 0, prove that r is not a partial limit
of (r
n
).
4. Let r
n
=
_
_
_
(1)
n
:
3
if : is a multiple of 3
0 if : is one more than a multiple of 3
4 if : is two more than a multiple of 3
. Prove
that the partial limits of (r
n
) are , , 0, 4.
5. Let r
n
=
3 + 2:
5 + :
. Prove that r
n
2.
6. Let r
n
=
_
_
1
2
n
if : = 2/
1
:
2
+ 1
if : = 2/ + 1
for any integer /. Prove that r
n

0.
7. Let r
n
=
:
2
+ 3: + 1
2:
2
+ : + 4
. Prove that r
n

1
2
.
8. Explain and justify the following observations made in the notes:
(a) If a sequence is in a set, then it is eventually in the set.
(b) If a sequence is eventually in a set, then it is frequently in the set.
(c) (r
n
) is eventually in o if and only if it is not frequently in R o.
(d) (r
n
) is frequently in o if and only if it is not eventually in R o.
9. Prove theorem 288.
4.3. LIMIT OF A SEQUENCE: THEOREMS 115
4.3 Limit of a Sequence: Theorems
These theorems fall in two categories. The rst category deals with ways to
combine sequences. Like numbers, sequences can be added, multiplied, divided,
... Theorems from this category deal with the ways sequences can be combined
and how the limit of the result can be obtained. If a sequence can be written
as the combination of several "simpler" sequences, the idea is that it should be
easier ti nd the limit of the "simpler" sequences. These theorems allow us to
write a limit in terms of easier limits. however, we still have limits to evaluate.
The second category of theorems deal with specic sequences and techniques
applied to them. Usually, computing the limit of a sequence involves using
theorems from both categories.
4.3.1 Limit Properties
We begin with a few technical theorems. They do not play an important role in
computing limits, but they play a role in proving certain results about limits.
Theorem 310 Let r be a number such that \c 0, [r[ < c, then r = 0.
Proof. See problems at the end of the section.
Theorem 311 If a sequence converges, then its limit is unique.
Proof. We do a proof by contradiction. Assume that a
n
1
1
and a
n
1
2
.
Given c 0 choose
1
such that : _
1
== [a
n
1
1
[ <
c
2
. Similarly, choose
2
such that : _
2
== [a
n
1
2
[ <
c
2
. Let = max (
1
,
2
). If : _ ,
then
[1
1
1
2
[ = [1
1
a
n
+ a
n
1
2
[
_ [a
n
1
1
[ +[a
n
1
2
[
<
c
2
+
c
2
= c
By theorem 310 , it follows that 1
1
1
2
= 0, that is 1
1
= 1
2
.
Theorem 312 If a sequence converges, then it is bounded, that is there exists
a number ' 0 such that [a
n
[ _ ' for all :.
Proof. Choose such that : _ ==[a
n
1[ < 1. By the triangle inequality,
we have
[a
n
[ [1[ _ [a
n
1[
< 1
Thus, if : _ , [a
n
[ < 1 + [1[. Let '
1
= max ([a
1
[ , [a
2
[ , ..., [a
N
[). Let ' =
max ('
1
, 1 +[1[). Then, clearly, [a
n
[ < '
Theorem 313 If a sequence converges, then \c 0 : :, : _ ==
[a
n
a
m
[ < c
Proof. Given c 0, we can choose such that :, : _ == [a
n
1[ <
c
2
and [a
m
1[ <
c
2
. Now,
[a
n
a
m
[ = [a
n
1 + 1 a
m
[
= [(a
n
1) (a
m
1)[
_ [(a
n
1)[ +[a
m
1[ by the triangle inequality
<
c
2
+
c
2
= c
The above theorem simply says that if a sequence converges, then the dif-
ference between any two terms gets smaller and smaller. It should also be clear
to the reader that if a
n
1, then so does a
n+k
where / is any natural number.
The above theorem can be used to prove that a sequence does not converge
by proving that the dierence between two of its terms does not get smaller and
smaller.
Example 314 Find limcos :
We suspect the sequence diverges, as its values will oscillate between -1 and
1 . We can actually prove it using theorem 313. We notice that for any :,
[cos 2: cos ((2: + 1) )[ = 2. For the sequence to converge, this dierence
should approach 0. Hence, the sequence diverges.
Theorem 315 Suppose that (a
n
) converges. Then, any subsequence (a
n
k
) also
converges and has the same limit.
Proof. Suppose that a
n
1. Let /
k
= a
n
k
be a subsequence. We need to prove
that given c 0, there exists such that / _ == [/
k
1[ < c. Let c 0
be given. Choose such that :
k
_ == [a
n
k
1[ < c. Now, if / _ , then
:
k
_ therefore
[/
k
1[ = [a
n
k
1[
< c
This theorem is often used to show that a given sequence diverges. To do so,
it is enough to nd two subsequences which do not converge to the same limit.
Alternatively, once can nd a subsequence which diverges.
Example 316 Study the convergence of cos :
The subsequence cos 2: converges to 1, while the subsequence cos (2: + 1)
converges to 1. Thus, cos 2: must diverge.
In the next two sections, we look at theorems which give us more tools to
compute limits.
4.3.2 Limit Laws
The theorems below are useful when nding the limit of a sequence. Finding the
limit using the denition is a long process which we will try to avoid whenever
possible. Since all limits are taken as : , in the theorems below, we will
write lima
n
for lim
n!1
a
n
.
Theorem 317 Let (a
n
) and (/
n
) be two sequences such that a
n
a and /
n
/
with a and / real numbers. Then, the following results hold:
1. lim(a
n
/
n
) = (lima
n
) (lim/
n
) = a /
2. lim(a
n
/
n
) = (lima
n
) (lim/
n
) = a/
3. if lim/
n
= / ,= 0 then lim
_
a
n
/
n
_
=
lima
n
lim/
n
=
a
/
4. lim[a
n
[ = [lima
n
[ = [a[
5. if a
n
_ 0 then lima
n
_ 0
6. if a
n
_ /
n
then lima
n
_ lim/
n
7. if lima
n
= a _ 0 then lim
_
a
n
=
_
lima
n
=
_
a
Proof. We prove some of these items. The remaining ones will be assigned as
problems at the end of the section.
1. We prove lim(a
n
+ /
n
) = (lima
n
)+(lim/
n
). The proof of lim(a
n
/
n
) =
(lima
n
) (lim/
n
) is left as an exercise. We need to prove that \c 0,
: : _ == [a
n
+ /
n
(a + /)[ < c. Let c 0 be given, choose
1
such that : _
1
== [a
n
a[ <
c
2
. Choose
2
such that : _
2
==
[/
n
/[ <
c
2
. Let = max (
1
,
2
). If : _ , then
[a
n
+ /
n
(a + /)[ = [(a
n
a) + (/
n
/)[
_ [a
n
a[ +[/
n
/[ by the triangle inequality
<
c
2
+
c
2
= c
2. We need to prove that \c 0, : : _ == [a
n
/
n
a/[ < c. Since
a
n
converges, it is bounded, let ' be the bound i.e. [a
n
[ < '. Choose
1
such that : _
1
== [a
n
a[ <
c
2 ([/[ + 1)
. Choose
2
such that
: _
2
==[/
n
/[ <
c
2 (' + 1)
. Let = max (
1
,
2
). If : _ then
[a
n
/
n
a/[ = [a
n
/
n
a
n
/ + a
n
/ a/[
= [a
n
(/
n
/) + / (a
n
a)[
_ [a
n
[ [/
n
/[ +[/[ [a
n
a[
< '
c
2 (' + 1)
+[/[
c
2 ([/[ + 1)
<
c
2
+
c
2
= c
3. See problems
4. We need to prove that [a
n
[ [a[ that is \c 0, : : _ ==
[[a
n
[ [a[[ < c. Let c 0 be given, choose such that : _ ==
[a
n
a[ < c (since a
n
a) . If : _ , then we have:
[[a
n
[ [a[[ < [a
n
a[ by the triangle inequality
< c
5. We prove it by contradiction. Assume that a
n
a < 0. Choose such
that : _ =[a
n
a[ <
1
2
a. Then,
a
n
a <
1
2
a
=a
n
<
1
2
a
=a
n
< 0
which is a contradiction.
6. We apply the results found in parts 1 and 5 to the sequence a
n
/
n
.
7. See problems
Remark 318 Parts 1, 2 and 3 of the above theorem hold even when a and /
are extended real numbers as long as the right hand side in each part is dened.
You will recall the following rules when working with extended real numbers:
1. += = () () =
2. = () = () =
3. If r is any real number, then
(a) + r = r +=
(b) + r = r =
(c)
r
=
r
= 0
(d)
r
0
=
_
if r 0
if r < 0
(e) r = r =
_
if r 0
if r < 0
(f ) () r = r () =
_
if r 0
if r < 0
4. However, the following are still indeterminate forms. Their behavior is
unpredictable. Finding what they are equal to requires more advanced tech-
niques such as lHpitals rule.
(a) + and
(b) 0 and 0
(c)

and
0
0
Remark 319 When using theorems from this category, it is important to re-
member previous results since these theorems allow us to write a limit in terms
of other limits, we hopefully know. The more limits we know, the better o we
are.
Example 320 If c ,= 0, nd lim
c
n
. We know from an example in the previous
section that lim
1
n
= 0. Therefore
lim
c
:
= c lim
1
:
= c 0
= 0
Example 321 Find lim
1
n
2
. In the previous section, we computed this limit
using the denition. We can also do it as follows.
lim
1
:
2
= lim
_
1
:

1
:
_
=
_
lim
1
:
__
lim
1
:
_
= 0 0
= 0
Remark 322 From the above example, we can see that if j is a natural number,
lim
1
n
p
= 0.
n
2
+3n
2n
2
+1
.
This problem involves using a standard technique you should remember. We
show all the steps, then we will draw a general conclusion. We begin by factoring
the term of highest degree from both the numerator and denominator.
lim
:
2
+ 3:
2:
2
+ 1
= lim
:
2
_
1 +
3
n
_
:
2
_
2 +
1
n
2
_
= lim
1 +
3
n
2 +
1
n
2
Now,
lim
_
1 +
3
:
_
= lim(1) + 3 lim
1
:
= 1
and
lim
_
2 +
1
:
2
_
= lim(2) + lim
1
:
2
= 2
Since both the limit of the numerator and denominator exist and the limit of the
denominator is not 0, we can write
lim
:
2
+ 3:
2:
2
+ 1
=
lim
_
1 +
3
n
_
lim
_
2 +
1
n
2
_
=
1
2
Remark 324 The same technique can be applied to every fraction for which
the numerator and denominator are polynomials in :. We see that the limit
of such a fraction will be the same as the limit of the quotient of the terms of
highest degree.
4.3.3 More Theorems on Limits
In example 304 , we used an approximation to simplify the problem a little bit.
In this particular example, the approximation was not really necessary, it was
more to illustrate a point. Sometimes, if the problem is more complicated, it
may be necessary to use such an approximation in order to be able to nd the
condition : has to satisfy. In other words, when we try to satisfy [r
n
1[ < c,
we usually simplify [r
n
1[ to some expression involving :. Let 1
1
(:) denote
this expression. This gives us the inequality 1
1
(:) < c which we have to solve
for :. If it is too hard, we then try to nd a second expression we will call 1
2
(:)
such that 1
1
(:) < 1
2
(:) < c. 1
2
(:) should be such that solving the inequality
1
2
(:) < c is feasible and easier. In order to achieve this, several tricks are used.
We recall some useful results, as well as some theorems below.
Theorem 325 (Bernoullis inequality) If r _ 1, and : is a natural num-
ber, then (1 + r)
n
_ 1 + :r.
Theorem 326 (binomial theorem) (a + /)
n
= a
n
+:a
n1
/+
:(: 1)
2
a
n2
/
2
+
:(: 1) (: 2)
3!
a
n3
/
3
+ ... + :a/
n1
+ /
n
.
Corollary 327 (1 + r)
n
= 1 + :r +
:(: 1)
2
r
2
+
:(: 1) (: 2)
3!
r
3
+ ... +
:r
n1
+ r
n
.
In particular, when r _ 0, then (1 + r)
n
is greater than any part of the right
hand side. For example, we obtain Bernoullis inequality: (1 + r)
n
_ 1 + :r.
We could also write (1 + r)
n
_
:(: 1)
2
r
2
or (1 + r)
n
_
:(: 1) (: 2)
3!
r
3
.
And so on. This is useful to get approximations on quantities like 3
n
. We
rewrite it as
3
n
= (1 + 2)
n
Theorem 328 (squeeze theorem) If a
n
1, c
n
1 and a
n
_ /
n
_ c
n
,
then /
n
1
Proof. We need to prove that \c 0 : : _ == [/
n
1[ < c. Let c 0
be given. Choose
1
such that : _
1
== [a
n
1[ < c or c < a
n
1 < c.
Similarly, choose
2
such that : _
2
== c < c
n
1 < c. Let =
max (
1
,
2
). If : _ then
a
n
_ /
n
_ c
n
==a
n
1 _ /
n
1 _ c
n
1
==c < a
n
1 _ /
n
1 _ c
n
1 < c
==c < /
n
1 < c
==[/
n
1[ < c
Theorem 329 If 0 < a < 1 then a
n
0
Proof. Let r =
1
a
1. Then, r 0 and a =
1
1 + r
. For : _ 1,
a
n
=
1
(1 + r)
n
_
1
1 + :r
by Bernoullis inequality
<
1
:r
To show that a
n
0, we need to show that [a
n
[ < c, for any c whenever : _ ,
that is
a
n
< c ==
1
:r
< c
==:
1
rc
So, given c 0, , =
1
rc
will work.
We look at a few more examples, to see how all these results come into play.
2:
2
+ 1
:
2
+ 1
using the denition
Of course, we can nd this limit by using the theorems on limits. Here, we do
it using the denition, as asked. We think the limit is 2. We want to show that
for every c 0, there exists such that : _ ==
2:
2
+ 1
:
2
+ 1
2
< c. First,
we simplify the absolute value.
2:
2
+ 1
:
2
+ 1
2
1
:
2
+ 1
=
1
:
2
+ 1
<
1
:
2
<
1
:
if : 1
So, we see that if
1
:
< c, which happens when :
1
c
, then we will have
2:
2
+ 1
:
2
+ 1
2
< c. So, =
1
c
will work.
1
:
2
+ 2: 4
We think the limit is 0. We need to prove that for every c 0, there exists
such that : _ ==
1
:
2
+ 2: 4
< c. We begin by noticing that

:
2
+ 2: 4 = :
2
+ 2 (: 2)
:
2
if : 2
Therefore, if : 2,
1
:
2
+ 2: 4
=
1
:
2
+ 2: 4
<
1
:
2
<
1
:
If : max
_
2,
1
c
_
, then
1
:
2
+ 2: 4
< c. So, = max

_
2,
1
c
_
will work.
Remark 332 The two examples above could have been done using the tech-
niques discussed in the previous subsection, that is without using the denition.
Example 333 Prove that lim
4
n
:!
= 0
The trick we use is worth remembering. We will use the squeeze theorem. We
note that
0 _
4
n
:!
=
4 4 4 ... 4
1 2 3 ... :
_
4
3
3!

4
:
since
4 4 ... 4
4 5 ... : 1
_ 1
But lim
4
3
3!

4
n
= 0 hence the result follows by the squeeze theorem.
4
n
2:
We suspect this sequence diverges to innity. We need to prove that
4
n
2:
grows
without bounds, that is for every ' 0, there exists such that : _ ==
4
n
2:
'. First, we notice that
4
n
2:
=
(1 + 3)
n
2:
:(: 1) 3
2
2
2:
by using the binomial theorem
9 (: 1)
4
We can see that
9 (: 1)
4
' ==: 1
4'
9
==:
4' + 9
9
so, =
4' + 9
9
will work.
Remark 335 We used the binomial theorem to deduce that (1 + 3)
n
_
:(: 1) 3
2
2
.
Often, when using this theorem, students do not know which term to keep, the
term in r, or r
2
, or r
3
,... The answer is that it depends on what we are trying
to achieve. Here, we wanted to show that
4
n
2:
' for any ', in other words,
4
n
2:
can be made as big as we want simply by taking : large enough. Since we
could not solve for :, we replaced
4
n
2:
by a smaller term. That smaller term
should have similar properties, in other words, the smaller term should also get
arbitrarily large. If we take the smaller term too small, it will not work. For
example, it is true that
4
n
2:

1
2:
. However,
1
2:
cannot be made as large as
we want, so this does not help us. If instead of using (1 + 3)
n
_
:(: 1) 3
2
2
(terms of second degree), we had used (1 + 3)
n
_ 1 + 3:, then we would have
obtained
1 + 3:
2:
' ==2:' < 1 + 3:
== 2:' 3: < 1
== :(2' 3) < 1
== : <
1
2' 3
So, this does not work. This tells us that
1+3n
2n
' only for a few values of :.
This is why we used terms of higher degree.
__
: + 2
_
:
_
lim
__
: + 2
_
:
_
= lim
__
: + 2
_
:
_ __
: + 2 +
_
:
_
__
: + 2 +
_
:
_
= lim
2
_
: + 2 +
_
:
=
2
lim
__
: + 2 +
_
:
_
= 0
Remark 337 In many proofs or problems, dierent versions of the triangle
inequality are often used. As a reminder, here are the dierent versions of the
triangle inequality students should remember.
[[a[ [/[[ _ [a /[ _ [a[ +[/[
and
[[a[ [/[[ _ [a + /[ _ [a[ +[/[
Finally, we give a theorem which generalizes some of the examples we did
above.
Theorem 338 All limits are taken as : .
1. If j 0 then lim
1
:
p
= 0.
2. If j 0 then lim
n
_
j = 1.
3. lim
n
_
: = 1.
4. If j 1 and c R then lim
:
j
n
= 0.
5. If [j[ < 1 then limj
n
= 0
6. \j R, lim
j
n
:!
= 0.
Proof. We prove each part separately.
1. We can use the denition. Let c 0 be given. We show there exists
0 : : _ ==
1
:
p
0
< c. We begin with what we want to achieve.
1
:
p
0
< c ==
1
:
p
< c
== :
p
1
c
== :
1
p
_
c
So, we see that if is the smallest ineteger larger than
1
p
p
, the result will

follow.
2. See problems
3. Let r
n
=
n
_
: 1. Proving the result amounts to proving that limr
n
= 0.
From r
n
=
n
_
: 1, we can write
n
_
: = r
n
+ 1
: = (r
n
+ 1)
n
Using the binomial theorem, we see that if : _ 2
: _
:(: 1)
2
r
2
n
Thus
r
2
n
_
2
: 1
It follows that for all : _ 2 we have
0 _ r
n
_
_
2
: 1
Since lim
_
2
n1
= 0, it follows using the squeeze theorem that limr
n
= 0.
4. Here again, we will use the binomial theorem. Since j 1, we can write
j = 1 + with 0. Therefore, if / is a positive integer such that / c,
we have
j
n
= (1 + )
n
:(: 1) ... (: / + 1)
/!

k
Now, if : 2/, then / <
1
2
:. It follows that : / +1
1
2
: +1
1
2
:. It
follows that
:(: 1) ... (: / + 1)
/!

:
k
2
k
/!
and therefore
0 _
:
j
n
_
2
k
/!
k

1
:
k
Since
lim
n!1
2
k
/!
k

1
:
k
=
2
k
/!
k
lim
n!1
1
:
k
= 0
It follows from the squeeze theorem that lim
:
j
n
= 0
5. See problems (hint: write j =
1
q
and use part 4 of the theorem with
c = 0).
6. See problems.
4.3.4 Exercises
1. Prove that
5
n
:!
0.
2. Prove that
:!
:
n
0.
3. Finish proving theorem 338.
4. Consider (r
n
) a sequence of real numbers such that lim
n!1
r
n
= r where
r 0, prove there exists an integer 0 such that : _ ==r
n
0.
5. Consider (r
n
) a sequence of real numbers such that r
n
_ 0 for any :. If
r is a partial limit of (r
n
), prove that r _ 0. Prove the same result if
r = limr
n
. Use this to show that if r
n
_ j
n
then limr
n
_ limj
n
.
6. Consider (r
n
) a sequence of real numbers and let r R. Prove the two
conditions below are equivalent.
(a) r
n
r as : .
(b) [r
n
r[ 0 as : .
7. Consider (r
n
) a sequence of real numbers and let r R. If limr
n
= r,
prove that lim[r
n
[ = [r[.
8. Consider (r
n
) a sequence of real numbers and let r R. Suppose that
limr
n
= r. Dene j
n
= r
n+p
for some integer j. Prove that limj
n
= r.
9. Given that a
n
_ /
n
for every : and that lima
n
= , prove that lim/
n
=
.
10. Suppose that (a
n
) and (/
n
) are sequences of real numbers such that
[a
n
/
n
[ < 1 for every :. Prove that if is a partial limit of (a
n
)
then is also a partial limit of (/
n
).
11. Two sequences are said to be eventually close if \c 0, 0 : : _
==[a
n
/
n
[ < c.
(a) Prove that if two sequences (a
n
) and (/
n
) are eventually close and if
a number r is the limit of the sequence (a
n
) then r is also the limit
of the sequence (/
n
).
(/) Prove that if two sequences (a
n
) and (/
n
) are eventually close and if
a number r is a partial limit of the sequence (a
n
) then r is also a
partial limit of the sequence (/
n
).
12. Determine if the sequences below have limits. In each case, if a limit 1
exists, given c 0, nd such that : _ ==[r
n
1[ < c.
(a)
: 1
: + 1
(b)
3:
2
+ 1
:
2
+ 1
(c)
5:
3
+ : 4
2:
3
+ 3
(d)
:
3
: + 1
(e)
4:
2
+ 1
:
3
+ :
(f)
1
:
2
+
2
:
2
+ ... +
:
:
2
(g) (1)
2n1
_
1
1
2
n
_
(h)
_
1
1
2
__
1
1
3
__
1
1
4
_
...
_
1
1
:
_
(i)
5
n
2:
(j) Prove that if a 1 then a
2
a
_
a 1
(k) Prove that
:
2
n
0,
:
2
2
n
0,
:
3
2
n
0,
:
4
2
n
0
13. Prove theorem 310
14. Finish proving theorem 317
15. Let (r
n
) and (j
n
) be two sequences. Assume further that limr
n
= r and
j is a partial limit of (j
n
). Prove that r+j is a partial limit for (r
n
+ j
n
).
16. State and prove similar results for subtraction, multiplication and division.
17. Give an example of two divergent sequences (r
n
) and (j
n
) such that
(r
n
+ j
n
) is convergent.
18. Give an example of two sequences (r
n
) and (j
n
) such that r
n
0, j
n

and:
(a) r
n
j
n
0
(b) r
n
j
n
c where 0 < c <
(c) r
n
j
n

(d) (r
n
j
n
) is bounded but has no limit.
19. Let (r
n
) be a sequence of real numbers such that r
n
0. Prove that
r
1
+ r
2
+ ... + r
n
:
0
20. Let (r
n
) be a sequence of real numbers such that r
n
r for some real
number r. Prove that
r
1
+ r
2
+ ... + r
n
:
r
21. Show by examples that if [r
n
[ [r[ then r
n
does not necessarily converge.
If it does converge, it does not necessarily converge to r.
22. Prove that [r
n
[ 0 ==r
n
0.
23. Show by examples that if (r
n
) and (j
n
) are divergent sequences then
(r
n
+ j
n
) is not necessarily divergent. Do the same for (r
n
j
n
)
24. Show that if (r
n
) 0 and (j
n
) is bounded then (r
n
j
n
) 0.
25. Prove that if (r
n
) is a sequence in a set o _ R, then every nite partial
limit of (r
n
) must belong to o.
26. Prove that if o _ R and r o then there exists a sequence (r
n
) in o such
that r
n
r.
27. Let o _ R. Prove that the following conditions are equivalent:
(a) o is unbounded above.
(b) There exists a sequence (r
n
) in o such that r
n
.
4.4. MONOTONE SEQUENCES AND CAUCHY SEQUENCES 129
4.4 Monotone Sequences and Cauchy Sequences
4.4.1 Monotone Sequences
The techniques we have studied so far require we know the limit of a sequence
in order to prove the sequence converges. However, it is not always possible
to nd the limit of a sequence by using the denition, or the limit rules. This
happens when the formula dening the sequence is too complex to work with.
It also happens with sequences dened recursively. Furthermore, it is often the
case that it is more important to know if a sequence converges than what it
converges to. In this section, we look at two ways to prove a sequence converges
without knowing its limit.
We begin by looking at sequences which are monotone and bounded. These
terms were dened at the beginning of this chapter.
You will recall that in order to show that a sequence is increasing, several
methods can be used.
1. Direct approach, simply show that a
n+1
_ a
n
for every :.
2. Equivalently, show that a
n+1
a
n
_ 0 for every :.
3. Equivalently, show that
a
n+1
a
n
_ 1 for every : if both a
n
and a
n+1
are
positive.
4. If a
n
= ) (:), one can show a sequence (a
n
) is increasing by showing that
) is increasing that is by showing that )
0
(r) _ 0.
5. By induction.
We now state and prove an important theorem about the convergence of
increasing sequences.
Theorem 339 An increasing sequence (a
n
) which is bounded above converges.
Furthermore, lim
n!1
a
n
= supa
n
.
Proof. We need to show that given c 0, there exists such that : _ ==
[a
n
[ < c where is the limit. Let c 0 be given. Since (a
n
) is bounded
above, by theorem 158, a
n
has a supremum. Let = supa
n
. Let c 0 be
given. then, c < . By theorem 151, there exists an element of a
n
, call
it a
N
, such that c < a
N
_ . Since (a
n
) is increasing, we have a
n
_ a
N
for every : _ . Therefore, if : _ ,
c < a
n
_ ==c < a
n
_ 0
==0 _ a
n
< c
==[a
n
[ < c
Corollary 340 A decreasing sequence (a
n
) which is bounded below converges.
Furthermore, lim
n!1
a
n
= inf a
n
.
Proof. The proof is similar to the proof of the theorem.
Theorem 341 A monotone sequence converges if and only if it is bounded.
Proof. The proof follows from results proven in the previous section and the
previous theorem and its corollary. Suppose we have a monotone sequence (r
n
).
If we assume that (r
n
) converges, then it follows that it is bounded by theorem
312 that it is bounded. Conversely, if we assume that (r
n
) is bounded then
it is both bounded above and below. Since (r
n
) is monotone, then it is either
increasing or decreasing. If it is increasing, it will converge by the previous
theorem since it is also bounded above. If it is decreasing, then it will converge
by the corollary since it is bounded below.
Example 342 Prove that the sequence whose general term is a
n
=
n
k=0
1
/!
con-
verges.
We try to establish this result by showing that this sequence is non-decreasing
and bounded above.
(a
n
) is increasing: a
n
= 1 + 1 +
1
2
+
1
3!
+ ... +
1
:!
. Therefore,
a
n+1
a
n
=
1
(: + 1)!
0
(a
n
) is bounded above. For this, we use the fact that :! _ 2
n1
for every
: _ 1. The proof of this fact is left as an exercise. Therefore,
1 + 1 +
1
2
+
1
3!
+ ... +
1
:!
< 1 + 1 +
1
2
+
1
2
2
+ ... +
1
2
n1
< 1 +
_
1
2
_
0
+
_
1
2
_
1
+
_
1
2
_
2
+ ... +
_
1
2
_
n1
< 1 +
1
_
1
2
_
n
1
1
2
< 1 +
1
1
1
2
< 3
Since (a
n
) is bounded above and increasing, it must converge. We will call
the limit c
0
.
n!1
a
n
where (a
n
) is dened by:
a
1
= 2
a
n+1
=
1
2
(a
n
+ 6)
If we knew the limit existed, nding it would be easy. We must rst establish that
it exists. We do this by showing that this sequence is increasing and bounded
above. This part is left as an exercise. Once this fact has been established, then
we know the sequence must converge. Let 1 be its limit, we must nd 1. Before
proceeding, we will recall the following fact: lima
n+1
= lima
n
. Therefore, if for
every : we have a
n+1
=
1
2
(a
n
+ 6), then we must also have:
lima
n+1
= lim
1
2
(a
n
+ 6)
1 =
1
2
(lima
n
+ 6)
1 =
1
2
(1 + 6)
21 = 1 + 6
1 = 6
Example 344 Let a
n
=
_
1 +
1
:
_
n
. Show lima
n
exists. This limit is in fact
the number c, but we wont show that. Again, to show that (a
n
) converges, we
show that it is increasing and bounded above.
(a
n
) bounded above. To establish this, we use the following fact: If / is
an integer such that 1 < / _ :, then
:(: 1) (: 2) ... (: / + 1)
:
k
=
_
1
1
:
__
1
2
:
_
...
_
1
/ 1
:
_
The proof of this is left as an exercise. Using the binomial theorem, we
have:
a
n
= 1 + :
1
:
+
:(: 1)
2!
_
1
:
_
2
+ ... +
:(: 1) (: 2) ... (: / + 1)
/!
_
1
:
_
k
+ ... +
:!
:!
_
1
:
_
n
= 1 + 1 +
1
2!
_
1
1
:
_
+ ... +
1
/!
_
1
1
:
__
1
2
:
_
...
_
1
/ 1
:
_
+ ... +
1
:!
_
1
1
:
__
1
2
:
_
...
_
1
: 1
:
_
< 1 + 1 +
1
2!
+ ... +
1
:!
< c
0
where c
0
is the limit found in the exercise above. Thus, (a
n
) is bounded by
c
0
.
(a
n
) increasing. The /t/ term in the expansion of a
n
is
1
/!
_
1
1
:
__
1
2
:
_
...
_
1
/ 1
:
_
.
The /t/ term in the expansion of a
n+1
is
1
/!
_
1
1
: + 1
__
1
2
: + 1
_
...
_
1
/ 1
: + 1
_
.
Since
,
: + 1
<
,
:
, it follows that 1
,
: + 1
1
,
:
. Hence, a
n+1
a
n
.
So, (a
n
) is increasing.
Since (a
n
) is increasing and bounded above, it follows that (a
n
) converges.
We dene this limit to be the number c.
4.4.2 Cauchy Sequences
Denition 345 (Cauchy Sequence) A sequence (r
n
) is said to be a Cauchy
sequence if for each c 0 there exists a positive integer such that :, : _
==[r
m
r
n
[ < c.
We begin with some remarks.
Remark 346 These series are named after the French mathematician Augustin
Louis Cauchy (1789-1857).
Remark 347 It is important to note that the inequality [r
m
r
n
[ < c must be
valid for all integers :, : that satisfy :, : _ . In particular, a sequence (r
n
)
satisfying [r
n+1
r
n
[ < c for all : _ may not be a Cauchy sequence.
Remark 348 A Cauchy sequence is a sequence for which the terms are even-
tually close to each other.
Remark 349 In theorem 313, we proved that if a sequence converged then it
had to be a Cauchy sequence. In fact, as the next theorem will show, there is a
stronger result for sequences of real numbers.
We now look at some examples.
Example 350 Consider (r
n
) where r
n
=
1
:
. Prove that this is a Cauchy
sequence.
Let c 0 be given. We want to show that there exists an integer 0 such
that :, : < ==[r
m
r
n
[ < c. That it we would like to have
[r
n
r
m
[ < c ==
1
:

1
:
< c
Since
1
:

1
:
<
1
:
+
1
:
If we make
1
:
+
1
:
< c, the result will follow. This will happen if both
1
:
<
c
2
that is when :
2
c
and
1
:
<
c
2
that is when :
2
c
. So, we see that if is an
integer larger than
2
c
then :, : ==[r
m
r
n
[ < c.
Example 351 Consider (r
n
) where r
n
=
n
k=1
1
/
2
Let c 0 be given. We want to show that there exists an integer 0 such
that :, : < ==[r
m
r
n
[ < c (without loss of generality, let us assume that
: :). That it we would like to have
[r
n
r
m
[ < c ==
k=1
1
/
2

m
k=1
1
/
2
< c
==
k=m+1
1
/
2
< c
Remembering telescoping sums and the fact that
1
/
2
<
1
/ (/ 1)
, we see that
[r
n
r
m
[ =
k=m+1
1
/
2
=
n
k=m+1
1
/
2
<
n
k=m+1
_
1
/ 1

1
/
_
=
1
:

1
:
<
1
:
+
1
:
So, as we saw in the previous example, if is an integer larger than
c
2
, then
:, : < ==[r
m
r
n
[ < c.
We now look at important properties of Cauchy sequences.
Theorem 352 Every Cauchy sequence is bounded.
Proof. See problems.
Theorem 353 A sequence of real numbers converges if and only if it is a
Cauchy sequence.
Proof. We have already proven one direction. Let (r
n
) be a sequence of real
numbers.
If (r
n
) converges, then we know it is a Cauchy sequence by theorem 313.
Assume (r
n
) is a Cauchy sequence. We must prove that it converges. By
theorem 352, we know that r
n
is bounded. Therefore, by the complete-
ness axiom, for each :, the number
a
n
= inf r
n
, r
n+1
, r
n+2
, ...
exists. Furthermore, the sequence (a
n
) is increasing and bounded (why?).
Therefore, by theorem 341, (a
n
) converges. Let a = lima
n
. We prove
that limr
n
= a. Let c 0 be given. Since (r
n
) is a Cauchy sequence, we
can nd 0 such that :, : _ == [r
n
r
m
[ <
c
2
. It follows that
both (r
n
) and (a
n
) are contained in the interval
_
r
N

c
2
, r
N
+
c
2
_
(why).
Thus, the limit a is also in this interval. Therefore, for all : _ , we
have:
[r
n
a[ _ [r
n
r
N
[ +[r
N
a[
< c
This proves that limr
n
= a, thus (r
n
) converges.
Remark 354 The key to this theorem is that we are dealing with a sequence of
real numbers. The fact that a Cauchy sequence of real number converges is linked
to the fact that R is complete. In fact, this is sometimes used as a denition of
completeness. Some texts say that a set is complete if every Cauchy sequence
converges in that set. It is possible to nd a Cauchy sequence of rational numbers
which does not converge in Q.
We nish this section with an important theorem.
Denition 355 A nested sequence of intervals is a sequence 1
n
of in-
tervals with the property that 1
n+1
_ 1
n
for all :.
Theorem 356 (Nested Intervals) If [a
n
, /
n
] is a nested sequence of closed
intervals then there exists a point . that belongs to all the intervals. Furthermore,
if lima
n
= lim/
n
then the point . is unique.
Proof. We provide a sketch of the proof and leave the details as homework.
First, prove that a
n
and /
n
are monotone and bounded. Thus, they must
converge. Let a = lima
n
and / = lim/
n
. Then, explain why a _ /. Conclude
from this.
4.4.3 Exercises
1. Prove that :! _ 2
n1
for every : _ 1
2. Prove that if / is an integer such that 1 < / _ :, then
:(: 1) (: 2) ... (: / + 1)
:
k
=
_
1
1
:
__
1
2
:
_
...
_
1
/ 1
:
_
3. Prove that the sequence given by
a
1
= 2
a
n+1
=
1
2
(a
n
+ 6)
is increasing and bounded above by 6. (hint: use induction for both).
4. Show that the sequence dened by
a
1
= 1
a
n+1
= 3
1
a
n
is increasing and satises a
n
< 3 for all :. Then, nd its limit.
5. Let a and / be two positive numbers such that a /. Let a
1
be their
arithmetic mean, that is a
1
=
a + /
2
. Let /
1
be their geometric mean, that
is /
1
=
_
a/. Dene a
n+1
=
a
n
+ /
n
2
and /
n+1
=
_
a
n
/
n
.
(a) Use mathematical induction to show that a
n
a
n+1
/
n+1
/
n
.
(b) Deduce that both (a
n
) and (/
n
) converge.
(c) Show that lima
n
= lim/
n
. Gauss called the common value of these
limits the arithmetic-geometric mean.
7. Answer the why? parts in the proof of theorem 353.
8. Finish proving theorem 356.

Sequences and Limit of Sequences: 4.1 Sequences: Basic de Nitions

Uploaded by

Document Information

Original Title

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

Sequences and Limit of Sequences: 4.1 Sequences: Basic de Nitions

Uploaded by

Copyright:

Available Formats

Chapter 4

Sequences and Limit of

< c, and we try to nd the

< c, and we try to nd the

< c. This time,

< c. The former

< c thus proving that

< c. We begin by noticing that

< c. So, = max

< c. We begin with what we want to achieve.

, the result will

You might also like