Sequenceof Setsand Limits

Download as pdf or txt
Download as pdf or txt
You are on page 1of 35

Measure Theory and Probability

Module II: Sequence of Sets and limits

Dr. Arindom Chakraborty

October 10, 2015


Objectives

I to define sequence of sets


Objectives

I to define sequence of sets


I to find limit superior (limsup) and limit inferior (liminf)
Objectives

I to define sequence of sets


I to find limit superior (limsup) and limit inferior (liminf)
I to find limit of a sequence of sets, if exists, using some
examples
Objectives

I to define sequence of sets


I to find limit superior (limsup) and limit inferior (liminf)
I to find limit of a sequence of sets, if exists, using some examples
I to define Indicator function of sets
Sequence and subsequence
I Sequence
Sequence and subsequence
I Sequence
Let N = {1, 2, ....., }. A sequence is a function f : N → R is
represented by
xn = f (n); n ∈ N (n ≥ 1)
Sequence and subsequence
I Sequence
Let N = {1, 2, ....., }. A sequence is a function f : N → R is
represented by
xn = f (n); n ∈ N (n ≥ 1)
A sequence can be given different ways
I List the elements.
{ 12 , 23 , 34 ....}
Sequence and subsequence
I Sequence
Let N = {1, 2, ....., }. A sequence is a function f : N → R is
represented by
xn = f (n); n ∈ N (n ≥ 1)
A sequence can be given different ways
I List the elements.
{ 12 , 23 , 34 ....}
I Give a formula to generate the terms.
n
xn = (−1)n 2n!
Sequence and subsequence
I Sequence
Let N = {1, 2, ....., }. A sequence is a function f : N → R is
represented by
xn = f (n); n ∈ N (n ≥ 1)
A sequence can be given different ways
I List the elements.
{ 12 , 23 , 34 ....}
I Give a formula to generate the terms.
n
xn = (−1)n 2n!
I A sequence can be given recursively.
x1 = 2; xn+1 = 21 (xn + 6)
Sequence and subsequence
I Sequence
Let N = {1, 2, ....., }. A sequence is a function f : N → R is
represented by
xn = f (n); n ∈ N (n ≥ 1)
A sequence can be given different ways
I List the elements.
{ 12 , 23 , 34 ....}
I Give a formula to generate the terms.
n
xn = (−1)n 2n!
I A sequence can be given recursively.
x1 = 2; xn+1 = 21 (xn + 6)
I Subsequence
Sequence and subsequence
I Sequence
Let N = {1, 2, ....., }. A sequence is a function f : N → R is
represented by
xn = f (n); n ∈ N (n ≥ 1)
A sequence can be given different ways
I List the elements.
{ 12 , 23 , 34 ....}
I Give a formula to generate the terms.
n
xn = (−1)n 2n!
I A sequence can be given recursively.
x1 = 2; xn+1 = 21 (xn + 6)
I Subsequence f : N → R; g : N → N 3 g (k) = nk ; k ≥ 1
with n1 ≤ n2 ≤ n3 ≤ ....
From the collection {xn ; n ≥ 1} we get a sub-collection
{xnk , k ≥ 1} with n1 ≤ n2 ≤ n3 ≤ ....
Subsequence of sets
A few examples,
I If Nn = {1, 2, 3, ..., n} then {Nn , n ≥ 1} is a sequence of subsets
of N
Subsequence of sets
A few examples,
I If Nn = {1, 2, 3, ..., n} then {Nn , n ≥ 1} is a sequence of subsets
of N
I For each n, define In = {x ∈ R; 0 < x < n1 } = (0, n1 ) is a
sequence of subsets of R
Subsequence of sets
A few examples,
I If Nn = {1, 2, 3, ..., n} then {Nn , n ≥ 1} is a sequence of subsets
of N
I For each n, define In = {x ∈ R; 0 < x < n1 } = (0, n1 ) is a
sequence of subsets of R
Subsequence of sets
A few examples,
I If Nn = {1, 2, 3, ..., n} then {Nn , n ≥ 1} is a sequence of subsets
of N
I For each n, define In = {x ∈ R; 0 < x < n1 } = (0, n1 ) is a
sequence of subsets of R
Union and Intersection of a Sequence of Subsets
Subsequence of sets
A few examples,
I If Nn = {1, 2, 3, ..., n} then {Nn , n ≥ 1} is a sequence of subsets
of N
I For each n, define In = {x ∈ R; 0 < x < n1 } = (0, n1 ) is a
sequence of subsets of R
Union and Intersection of a Sequence of Subsets
I Results about finite intersection and union of sets remain true in
this setting
Suppose {An } is a sequence of sets. We say
Subsequence of sets
A few examples,
I If Nn = {1, 2, 3, ..., n} then {Nn , n ≥ 1} is a sequence of subsets
of N
I For each n, define In = {x ∈ R; 0 < x < n1 } = (0, n1 ) is a
sequence of subsets of R
Union and Intersection of a Sequence of Subsets
I Results about finite intersection and union of sets remain true in
this setting
Suppose {An } is a sequence of sets. We say
I {An } is monotonically increasing (An ↑ n) if we have An ⊂ An+1
for every n in the domain of the sequence
e.g. An = [0, n]
I {An } is monotonically decreasing (An ↓ n) if we have An ⊃ An+1
for every n in the domain of the sequence
e.g. An = [0, n1 )
Limit superior and Limit inferior
For any sequence of sets {An } define
Limit superior and Limit inferior
For any sequence of sets {An } define
I
lim sup An = ∩∞ ∞
n=1 ∪k=n Ak
n
I
lim inf An = ∪∞ ∞
n=1 ∩k=n Ak
n

Define
I Bn = ∪∞
k=n Ak and

I Cn = ∩∞
k=n Ak
Note that Bn and Cn are both monotonic. In particular, Bn ↓ n
and Cn ↑ n.

B = lim Bn =
n
Limit superior and Limit inferior
For any sequence of sets {An } define
I
lim sup An = ∩∞ ∞
n=1 ∪k=n Ak
n
I
lim inf An = ∪∞ ∞
n=1 ∩k=n Ak
n

Define
I Bn = ∪∞
k=n Ak and

I Cn = ∩∞
k=n Ak
Note that Bn and Cn are both monotonic. In particular, Bn ↓ n
and Cn ↑ n.

B = lim Bn = ∩∞ ∞ ∞
n=1 Bn = ∩n=1 ∪k=1 Ak : lim supAn
n n
C = lim Cn =
n
Limit superior and Limit inferior
For any sequence of sets {An } define
I
lim sup An = ∩∞ ∞
n=1 ∪k=n Ak
n
I
lim inf An = ∪∞ ∞
n=1 ∩k=n Ak
n

Define
I Bn = ∪∞
k=n Ak and

I Cn = ∩∞
k=n Ak
Note that Bn and Cn are both monotonic. In particular, Bn ↓ n
and Cn ↑ n.

B = lim Bn = ∩∞ ∞ ∞
n=1 Bn = ∩n=1 ∪k=1 Ak : lim supAn
n n
C = lim Cn = ∪∞ ∞ ∞
n=1 Cn = ∪n=1 ∩k=n Ak : lim infAn
n n
Limit superior and Limit inferior
For any sequence of sets {An } define
I
lim sup An = ∩∞ ∞
n=1 ∪k=n Ak
n
I
lim inf An = ∪∞ ∞
n=1 ∩k=n Ak
n

Define
I Bn = ∪∞
k=n Ak and

I Cn = ∩∞
k=n Ak
Note that Bn and Cn are both monotonic. In particular, Bn ↓ n
and Cn ↑ n.

B = lim Bn = ∩∞ ∞ ∞
n=1 Bn = ∩n=1 ∪k=1 Ak : lim supAn
n n
C = lim Cn = ∪∞ ∞ ∞
n=1 Cn = ∪n=1 ∩k=n Ak : lim infAn
n n
A few examples
I Let An = (0, 1] or (1, 2) according to n is even or odd.
Then limn sup An = (0, 2) 6= limn inf An = φ
I An = {(x, y ) : 0 ≤ x < n, 0 ≤ y < n1 }, n ≥ 1
A few examples
I Let An = (0, 1] or (1, 2) according to n is even or odd.
Then limn sup An = (0, 2) 6= limn inf An = φ
I An = {(x, y ) : 0 ≤ x < n, 0 ≤ y < n1 }, n ≥ 1

1
lim sup An = ∩∞ ∞
n=1 ∪k=n {(x, y ) : 0 ≤ x < k, 0 ≤ y < }
n k
A few examples
I Let An = (0, 1] or (1, 2) according to n is even or odd.
Then limn sup An = (0, 2) 6= limn inf An = φ
I An = {(x, y ) : 0 ≤ x < n, 0 ≤ y < n1 }, n ≥ 1

1
lim sup An = ∩∞ ∞
n=1 ∪k=n {(x, y ) : 0 ≤ x < k, 0 ≤ y < }
n k
1
= ∩∞
n=1 {(x, y ) : 0 ≤ x < ∞, 0 ≤ y < }
n
= {(x, y ) : 0 ≤ x < ∞, y = 0}
A few examples
I Let An = (0, 1] or (1, 2) according to n is even or odd.
Then limn sup An = (0, 2) 6= limn inf An = φ
I An = {(x, y ) : 0 ≤ x < n, 0 ≤ y < n1 }, n ≥ 1

1
lim sup An = ∩∞ ∞
n=1 ∪k=n {(x, y ) : 0 ≤ x < k, 0 ≤ y < }
n k
1
= ∩∞
n=1 {(x, y ) : 0 ≤ x < ∞, 0 ≤ y < }
n
= {(x, y ) : 0 ≤ x < ∞, y = 0}

1
lim inf An = ∪∞ ∞
n=1 ∩k=n {(x, y ) : 0 ≤ x < k, 0 ≤ y < }
n k
A few examples
I Let An = (0, 1] or (1, 2) according to n is even or odd.
Then limn sup An = (0, 2) 6= limn inf An = φ
I An = {(x, y ) : 0 ≤ x < n, 0 ≤ y < n1 }, n ≥ 1

1
lim sup An = ∩∞ ∞
n=1 ∪k=n {(x, y ) : 0 ≤ x < k, 0 ≤ y < }
n k
1
= ∩∞
n=1 {(x, y ) : 0 ≤ x < ∞, 0 ≤ y < }
n
= {(x, y ) : 0 ≤ x < ∞, y = 0}

1
lim inf An = ∪∞ ∞
n=1 ∩k=n {(x, y ) : 0 ≤ x < k, 0 ≤ y < }
n k
= ∪∞
n=1 {(x, y ) : 0 ≤ x < n, y = 0}
= {(x, y ) : 0 ≤ x < ∞, y = 0}
A few examples
I Let An = (0, 1] or (1, 2) according to n is even or odd.
Then limn sup An = (0, 2) 6= limn inf An = φ
I An = {(x, y ) : 0 ≤ x < n, 0 ≤ y < n1 }, n ≥ 1

1
lim sup An = ∩∞ ∞
n=1 ∪k=n {(x, y ) : 0 ≤ x < k, 0 ≤ y < }
n k
1
= ∩∞
n=1 {(x, y ) : 0 ≤ x < ∞, 0 ≤ y < }
n
= {(x, y ) : 0 ≤ x < ∞, y = 0}

1
lim inf An = ∪∞ ∞
n=1 ∩k=n {(x, y ) : 0 ≤ x < k, 0 ≤ y < }
n k
= ∪∞
n=1 {(x, y ) : 0 ≤ x < n, y = 0}
= {(x, y ) : 0 ≤ x < ∞, y = 0}

Here lim sup An = lim inf An i.e. limit exists


n n
A few more examples
I Let An = (0, 1 − n1 )
1
lim inf An = ∪∞ ∞
n=1 ∩k=n (0, 1 − )
n k
= (0, 1) = lim sup An (Check!)
n

I In general,
lim inf An ⊂ lim sup An
n n

I Proof: We know

∩∞ ∞ ∞ ∞
k=n Ak ⊂ ∩k=n+j Ak ⊂ ∪k=n+j Ak ⊂ ∪k=j Ak

That is for any n, j

∩∞ ∞ ∞ ∞ ∞ ∞
k=n Ak ⊂ ∪k=j Ak ⇒ ∪n=1 ∩k=n Ak ⊂ ∩n=1 ∪k=n Ak

i.e. limn inf An ⊂ limn sup An


An important Theorem

IA monotone increasing (decreasing) sequence {An } is convergent


and lim An = ∪∞ ∞
n=1 An (∩n=1 An )
An important Theorem

IA monotone increasing (decreasing) sequence {An } is convergent


and lim An = ∪∞ ∞
n=1 An (∩n=1 An )

I Proof: If An ↑ n (i.e., is monotone increasing)

lim sup An = ∩∞ ∞ ∞ ∞
n=1 (∪k=n Ak ) = ∩n=1 (∪k=1 Ak )
n

since ∪∞ ∞ ∞
k=1 Ak = ∪k=n Ak (An ↑). But ∪k=1 Ak does not depend on
n and thus
lim sup An = ∪∞
n=1 An
n
An important Theorem

IA monotone increasing (decreasing) sequence {An } is convergent


and lim An = ∪∞ ∞
n=1 An (∩n=1 An )

I Proof: If An ↑ n (i.e., is monotone increasing)

lim sup An = ∩∞ ∞ ∞ ∞
n=1 (∪k=n Ak ) = ∩n=1 (∪k=1 Ak )
n

since ∪∞ ∞ ∞
k=1 Ak = ∪k=n Ak (An ↑). But ∪k=1 Ak does not depend on
n and thus
lim sup An = ∪∞
n=1 An
n

Also

lim inf An = ∪∞ ∞ ∞ ∞
n=1 (∩k=n Ak ) = ∪n=1 An since ∩k=n Ak = An
n
lim inf An = ∪∞
n=1 An = lim sup An
n n
Indicator function of Sets
(
1 ifw ∈ A
IA (w ) =
0 ifw ∈ AC

I IAC (w ) = 1 − IA (w )
I IA∩B = IA (w ).IB (w )
I IA∪B ≤ IA (w ) + IB (w )

Define
n
X n X
X n n
X n
X n
X
S1 = IAj , S2 = IAj1 Aj2 , .., Sm = .. IAj1 Aj2 ..Ajm
j=1 j1 =1 <j2 =1 j1 =1 <j2 =1< <jm =1

Sn = IAj1 Aj2 ..Ajn


A few results
I Result:
I∪ni=1 Ai = S1 − S2 + S3 − ...... + (−1)n−1 Sn

I / ∪i Ai ⇔ w ∈ ACi ∀i.
Proof: a). If w ∈
LHS=0; RHS=0
b). w ∈ ∪ni=1 Ai ⇔w ∈ Ai for i = 1..m say ,1 ≤ m < n
LHS=1; RHS= m1 − m2 + .... + (−1)m m m =1

I Result:
   
m+1 m+2
IBm = Sm − Sm+1 + Sm+2 − ...... + (−1)n−m Sn
m m
where Bm = Set of points belonging to exactly m sets
I Proof: Let us assume w ∈ exactly k sets viz A1 , A2 ...., Ak
If k < m, contribution to both LHS and RHS are zero.
If k = m, contribution to LHS=1, to RHS=1 (only Sm )
If k > m?

You might also like