National University of Singapore Department of Mathematics MA2216/ST2131 Probability Solution Set

NATIONAL UNIVERSITY OF SINGAPORE
DEPARTMENT OF MATHEMATICS
Year 2016/17
MA2216/ST2131 Probability
Tut. 1
Solution Set
1. Sol. The sample space S is the union of the following two sets:
{(x1 , x2 , . . . , xk ) : xi = 6 for i = 1, 2, . . . , k 1; xk = 6, k = 1, 2, . . .}
and
{(x1 , x2 , . . . , xn , . . .) : n, xn = 6}.
Obviously, En = {(x1 , x2 , . . . , xn ) : xi = 6 for i = 1, 2, . . . , n 1; xn = 6}. That
is, the sample points in En are those of which the nth roll results in an occurrence
of number 6 for the rst time. Observe also that
En = {(x1 , x2 , . . . , xk ) : xi = 6 for i = 1, 2, . . . , k 1; xk = 6, k = 1, 2, . . .},
which consists of all sample points with nite length.

( )c
For (iii),
En
= {(x1 , x2 , . . . , xn , . . .) : n, xn = 6}. Take note that En s are
1
mutually disjoint.
c
Now, assume that the die is a fair one. To evaluate IP {(
1 En ) }, one can nd
IP (En ) for each n rst, and then make use of the following observation:
{( )c }
{ }
En
En = 1
IP
= 1 IP
IP (En ).
1
(Here we make use of the -additive property.)

Clearly, for each n,
( )n1
5
1
.
IP (En ) =
6
6
(
)
(
)n
5 n1 1
1 5
1
1
IP (En ) =
=
=
6
6
6
6
6 1
n1
n1
n0
{( )c }
We conclude that IP
En
= 0.
5
6
= 1.
Finally, drop the fairness assumption and put p = IP {6} and q = 1 p = IP{non-6}.
Then, IP (En ) = q n1 p. There are three possible cases:
1
(a) If 0 < p < 1,
IP (En ) =
n1
q n1 p = p
n1
qn = p
n0
1
= 1.
1q
c
So, IP {(
1 En ) } = 0.
c
(b) If p = 0, then En = for all n, and hence, (
1 En ) = S, the sample space
c
itself. Thus, IP {(1 En ) } = 1.
c
(c) If p = 1, then S = {(6)} = E1 . In turn, it implies that (
1 En ) = and hence
c
IP {(
1 En ) } = 0.
We are done.
Remarks. In this exercise, one is concerned with the occurrence of 6, and hence,
one may classify the occurrence of 6 to be success. The occurrence of any non6 is then regarded as a failure. Thus, sample points can now be represented as
follows: (s), (f, s), (f, f, s), etc. In
other)words, E1 = {(s)}, E2 = {(f, s)},
(
c
E3 = {(f, f, s)}, etc., and obviously,

En
= {(f, f, f, . . .)}. By the way, this
1
has to do with the so-called Geometric model to be introduced soon in lecture.

For (iv), refer to Example 1 in 5.1, Notes 2.
2. Sol. An easy way to see a derivation is to draw a Veins diagram. The event that
exactly one of the events E or F occurs can be expressed as
(E \ F ) (F \ E)
and note that (E \F ) and (F \E) are mutually disjoint. So, the desired probability
is equal to
IP (E \ F ) + IP (F \ E).
Moreover, IP (E \ F ) = IP (E) IP (EF ) and IP (F \ E) = IP (F ) IP (EF ). We are
done.
Note. When A is a sub-event of B, i.e., A B, then IP(B \ A) = IP(B) IP(A).
To establish it, observe that B = (B \ A) A, of which (B \ A) and A are mutually
disjoint. Make use of the additive property of IP and we are done.
However, if A is not a sub-event of B, then the formula IP(B \ A) = IP(B) IP(A)
is not valid, in general.
(3)
3. Sol. Obviously, the desired probability is given by (n2 ) .
(3)
1
(n2 ) =
2
2
2
is equivalent to n (n 1) = 12. Thus, n = 4.

4. Sol. Let U , C, and T be events that the customer purchases a suit, shirt and a
tie, respectively. Then,
IP(U C T ) = IP(U ) + IP(C) + IP(T )
IP(U C) IP(U T ) IP(CT ) + IP(U CT )
= .22 + .30 + .28 .11 .14 .10 + .06
= .51.
The desired probability is then equal to 1 .51 = .49, which is the answer to (a).
For (b), we may draw a Venn diagram, which could help us see the event concerned
easier. Alternatively, observe rst that the probability of buying a suit only is given
by
IP{U \ (C T )} = IP(U ) IP(U C) IP(U T ) + IP(U CT )
= .22 .11 .14 + .06
= .03.
Likewise, the probability of purchasing the shirt only is given by
IP{C \ (U T )} = .30 .11 .10 + .06 = .15,
and IP{T \ (U C)} = .28 .14 .10 + .06 = .10.
Note that events U \ (C T ), C \ (U T ) and T \ (U C) are mutually disjoint.
The desired answer is the probability of the union of these three events, which is
thus equal to .03 + .15 + .10 = .28.
5. Sol. Clearly, the sample space can be expressed as
S = {(x1 , x2 ) : xi = 1, 2, , 6, i = 1, 2}
with |S| = 62 , of which each sample point is equally likely to occur. Put Ei to be the
event that the sum of the upturned faces equals i. It is sucient for us to determine
|Ei | for each i, and
|Ei |
IP(Ei ) =
.
|S|
(Refer to 2 of Notes 2.) Indeed,
i
|Ei |
i
|Ei |
2
1
3
2
8
5
9
4
3
4
3
10
3
5
4
11
2
6
5
7
6
12
1
6. Sol. First, one has to formulate the exercise in a mathematical way properly.
Then, make use of the inclusion-exclusion principle. Let A1 be the event that a
bridge hand is void in the suit spade, A2 void in heart, A3 void in diamond, and A4
void in club. We have to nd IP(A1 A2 A3 A4 ), which is equal to
IP(A1 ) + IP(A2 ) + IP(A3 ) + IP(A4 )
IP (A1 A2 ) IP (A1 A3 ) IP (A1 A4 ) IP (A2 A3 ) IP (A2 A4 ) IP (A3 A4 )
+ IP (A1 A2 A3 ) + IP (A1 A2 A4 ) + IP(A1 A3 A4 ) + IP (A2 A3 A4 )
IP(A1 A2 A3 A4 ).
(a) First, it is trivial to observe that A1 A2 A3 A4 = and hence, IP(A1 A2 A3 A4 ) = 0.
(b) A1 A2 A3 is the event that a bridge hand consists
(13) of only clubs. One can easily
)
compute its probability, which is equal to (13
52 . Moreover, it is trivial to note
13
that A1 A2 A3 , A1 A2 A4 , A1 A3 A4 and A2 A3 A4 are of equal probability.

(26)
)
(c) As for Ai1 Ai2 , its probability is given by (13
52 .
13
(39)
13
).
(d) Finally, IP(Ai ) = (52
13
( ) (39) ( ) (26) ( ) (13)

4
4
4
13
)
)+
).
Thus, the desired probability is
(13
(13
(52
52
52
1
2
3
13
13
13
7. Sol. First of all, an analytic derivation. Multiplying both sides by 2, the identity
becomes
n (n 1) = k (k 1) + 2 k (n k) + (n k) (n k 1).
Simplify the right side and obviously, it equals the left side. We are done.
Now, a combinatorial argument. Consider
(nan
) experiment of selecting two items from
a set S with |S| = n. Clearly, there are 2 ways of doing the experiment.
Let A S be any subset of k items.
Obviously, there are three dierent cases as described below: both of the 2 selected
items could be contained in A, or one of them belongs to A and the other belongs
to Ac , or both of them are contained in Ac .
(k )
2 is the number of ways that the 2 selected items are(contained
) in A; k (n k) is
nk
the number that only one item is selected from A; and
is the number that
2
neither is selected from A.
( )
Adding these three gives the total number, n .
2
8. Sol. Choose a committee of any size (size 1) with a chairperson:

(i) Method 1: Observe that there are n ways to choose a chairperson, and each
of the rest of n 1 people is either in or out of the committee. Thus, the total
number of ways of following this method is n 2n1 .
(ii) Method 2: Consider, for each k = 1, 2, . . . , n, forming a committee
of size k
( )
n
with a chairperson. To form a committee of size k, there are
ways. From
k
the committee formed, we then choose a chairperson, and there are k ways.
Obviously, the total
( )number of ways to form size k committee with an elected
n
chairperson is k
.
k
Consequently, for method 2, the number of ways of forming any size committee
( )
n
n
with an elected chairperson is
k
.
k
k=1
As the two ways of counting these committees must agree, we conclude that
n
k=1
( )
n
k
= n 2n1 .
k
(8.1)
Note. One may derive (8.1) by dierent ways. First, one may make use of the
mathematical induction on n. Another method is to consider the binomial identity
n ( )
n k
(1 + x) =
x
k
n
k=0
which is a function of x. Now dierentiate both sides with respect to x, etc. Try to
complete the derivation by yourself.
(
)
n+m
9. Sol. Refer to Ex. 10 below. Obviously, the answer is given by
.
r
(
)
n+m
10. Sol. Refer to Ex. 9. Observe that
is the number of groups of size r, and
r
( )(
)
n
m
for each i = 0, 1, . . . , r,
refers to the number of groups of size r that
i
ri
consists of i men and r i women. Such an observation establishes the identity.
11. Sol. An immediate consequence of Ex. 10 when both m and r in Ex. 10 are replaced
by n.
12. Sol. The procedures described in (i), (ii) and (iii), respectively, provide a combinatorial argument.
5
13. Sol. First, the solution to (i): for the rst digit, there are 9 ways (for 0 can not be
put there), for the 2nd digit, there are also 9 ways (as what has been put in the rst
digit can not be re-used again at this stage), etc. In the end, there are altogether
9n such numbers.
(
)
(ii) As number 0 can not be the leading digit for an n-digit number, there are n1
i
ways for i zeros to be placed among the remaining n 1 positions; as for the
rest of n i places, we may choose 1 to 9. So answer is
(
)
n 1 ni
9 .
i
( )
14. Sol. It amounts to choosing 2 out of 20, and hence the answer is given by 20
2 .
Alternatively, one may re-phrase it in the following way: say, there are n distinct
points, p1 , p2 , . . ., pn , on a circle. Now, connect every two (distinct) points with a
line segment. How many such line segments are there in all? Well, for p1 , there are
n 1 such segments to connect p1 with any other points. Nect, turn to p2 . Connect
it with a line segment with any other points among p3 , . . ., pn . (Why do we exclude
p1 here?) And, there are n 2 line segments. Repeat the same procedure till we
reach pn1 , and at that moment, there is only one line segment to connect points
pn1 and pn . Now, add up:
( )
(n 1) (1 + n 1)
n
(n 1) + (n 2) + + 2 + 1 =
=
.
2
2
15. Sol. One may, rst of all, choose j out of n people to form a committee of size j,
and then out of these
j people, choose i (i j) to form a subcommittee of
(n) (chosen
)
j
size i. There are j i of ways of doing it.
Alternatively, one may rst choose i out of n people to form a committee of size
i, and then out of the remaining n i people, choose further j i people and add
them to the committee just formed
to make
it a committee of size j. The number
(n) (ni
)
of ways of doing it is equal to i ji . Thus
( ) ( ) ( ) (
)
n
j
n
ni
(15.1)
j
i
i
ji
is established, which settles part (i).
For part (ii), observe that
)
n ( ) ( )
n ( ) (
n
j
n
ni
j
i
i
ji
j=i
j=i
(letting k = j i)
)
ni ( ) (
n
ni
=
(making use of 3.11(e), Notes 1)

i
k
k=0
( )
) ( )
ni (
n
ni
n ni
=
=
2 .
i
k
i
k=0
Finally, turn to part (iii). Observe that

)
n ( ) ( )
n ( ) (
n
j
n
ni
nj
(1)
=
(1)nj
j
i
i
j
i
j=i
j=i
(letting k = j i)
)
ni ( ) (
n
ni
(1)nik
=
i
k
k=0
( )
)
ni (
ni
n
=
(1)nik
i
k
k=0
( )
n
=
[1 + (1)]ni
i
= 0.

National University of Singapore Department of Mathematics MA2216/ST2131 Probability Solution Set

Uploaded by

Copyright:

Available Formats

National University of Singapore Department of Mathematics MA2216/ST2131 Probability Solution Set

Uploaded by

Document Information

Original Description:

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

National University of Singapore Department of Mathematics MA2216/ST2131 Probability Solution Set

Uploaded by

Copyright:

Available Formats

NATIONAL UNIVERSITY OF SINGAPORE

En = {(x1 , x2 , . . . , xk ) : xi = 6 for i = 1, 2, . . . , k 1; xk = 6, k = 1, 2, . . .},

which consists of all sample points with nite length.

(Here we make use of the -additive property.)

(a) If 0 < p < 1,

E3 = {(f, f, s)}, etc., and obviously,

has to do with the so-called Geometric model to be introduced soon in lecture.

is equivalent to n (n 1) = 12. Thus, n = 4.

that A1 A2 A3 , A1 A2 A4 , A1 A3 A4 and A2 A3 A4 are of equal probability.

( ) (39) ( ) (26) ( ) (13)

8. Sol. Choose a committee of any size (size 1) with a chairperson:

(making use of 3.11(e), Notes 1)

Finally, turn to part (iii). Observe that

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.