TUTORIAL 5

Chapter 4: Continuous R.V. (Part I) & Assignment

STAT1306 Introductory Statistics
(First Semester, 2011-2012)
Jun FU
Department of Statistics and Actuarial Science
The University of Hong Kong
October 10, 2011
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 1 / 37
Review of Key Concepts
Game Plan
1
Review of Key Concepts
Basic Denitions
Comparison Between Discrete and Continuous Random Variables
2
Problems
Continuous Random Variables
Some Questions from Assignments
3
Ending
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 2 / 37
Review of Key Concepts Basic Denitions
Game Plan
1
Review of Key Concepts
Basic Denitions
Comparison Between Discrete and Continuous Random Variables
2
Problems
Continuous Random Variables
Some Questions from Assignments
3
Ending
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 3 / 37
Review of Key Concepts Basic Denitions
Denitions
Continuous Random Variables: It can take any value in an interval.
Cumulative Distribution Function (c.d.f.):
F
Y
(y) = P(Y y) = P(Y < y).
Probability Density Function (p.d.f.):
f
Y
(y) =
dF
Y
(y)
dy
= F

Y
(y), F
Y
(y) =

f
Y
(s)ds.
It has the following properties:
(a) f
Y
(y) 0 for any value of y.
(b)

f
Y
(y)dy = F
Y
() = 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 4 / 37
Review of Key Concepts Basic Denitions
Denitions
Continuous Random Variables: It can take any value in an interval.
Cumulative Distribution Function (c.d.f.):
F
Y
(y) = P(Y y) = P(Y < y).
Probability Density Function (p.d.f.):
f
Y
(y) =
dF
Y
(y)
dy
= F

Y
(y), F
Y
(y) =

f
Y
(s)ds.
It has the following properties:
(a) f
Y
(y) 0 for any value of y.
(b)

f
Y
(y)dy = F
Y
() = 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 4 / 37
Review of Key Concepts Basic Denitions
Denitions
Continuous Random Variables: It can take any value in an interval.
Cumulative Distribution Function (c.d.f.):
F
Y
(y) = P(Y y) = P(Y < y).
Probability Density Function (p.d.f.):
f
Y
(y) =
dF
Y
(y)
dy
= F

Y
(y), F
Y
(y) =

f
Y
(s)ds.
It has the following properties:
(a) f
Y
(y) 0 for any value of y.
(b)

f
Y
(y)dy = F
Y
() = 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 4 / 37
Review of Key Concepts Basic Denitions
Denitions
Continuous Random Variables: It can take any value in an interval.
Cumulative Distribution Function (c.d.f.):
F
Y
(y) = P(Y y) = P(Y < y).
Probability Density Function (p.d.f.):
f
Y
(y) =
dF
Y
(y)
dy
= F

Y
(y), F
Y
(y) =

f
Y
(s)ds.
It has the following properties:
(a) f
Y
(y) 0 for any value of y.
(b)

f
Y
(y)dy = F
Y
() = 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 4 / 37
Review of Key Concepts Basic Denitions
Denitions
Continuous Random Variables: It can take any value in an interval.
Cumulative Distribution Function (c.d.f.):
F
Y
(y) = P(Y y) = P(Y < y).
Probability Density Function (p.d.f.):
f
Y
(y) =
dF
Y
(y)
dy
= F

Y
(y), F
Y
(y) =

f
Y
(s)ds.
It has the following properties:
(a) f
Y
(y) 0 for any value of y.
(b)

f
Y
(y)dy = F
Y
() = 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 4 / 37
Review of Key Concepts Comparison Between Discrete and Continuous Random Variables
Game Plan
1
Review of Key Concepts
Basic Denitions
Comparison Between Discrete and Continuous Random Variables
2
Problems
Continuous Random Variables
Some Questions from Assignments
3
Ending
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 5 / 37
Review of Key Concepts Comparison Between Discrete and Continuous Random Variables
Discrete R.V. vs Continuous R.V.
Basic Denition Discrete Continuous
P(Y A) =

yA
p(y)

A
f
Y
(y)dy
F
Y
(y) = P(Y y) P(Y y) = P(Y < y)
E(Y) =

y
y p(y)

y f
Y
(y)dy
E [g(Y)] =

y
g(y) p(y)

g(y) f
Y
(y)dy
Var (Y) = E
_
(Y )
2

y
(y )
2
p(y)

(y )
2
f
Y
(y)dy
M
Y
(t) = E
_
e
tY

y
e
ty
p(y)

e
ty
f
Y
(y)dy
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 6 / 37
Problems
Game Plan
1
Review of Key Concepts
Basic Denitions
Comparison Between Discrete and Continuous Random Variables
2
Problems
Continuous Random Variables
Some Questions from Assignments
3
Ending
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 7 / 37
Problems Continuous Random Variables
Game Plan
1
Review of Key Concepts
Basic Denitions
Comparison Between Discrete and Continuous Random Variables
2
Problems
Continuous Random Variables
Some Questions from Assignments
3
Ending
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 8 / 37
Problems Continuous Random Variables
For a function
f (x) =
_
( 1)x

x 1
0 x < 1
,
where > 1.
(a) Is f (x) a density function? Why?
(b) Find F(x).
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 9 / 37
Problems Continuous Random Variables
For a function
f (x) =
_
( 1)x

x 1
0 x < 1
,
where > 1.
(a) Is f (x) a density function? Why?
(b) Find F(x).
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 9 / 37
Problems Continuous Random Variables
For a function
f (x) =
_
( 1)x

x 1
0 x < 1
,
where > 1.
(a) Is f (x) a density function? Why?
(b) Find F(x).
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 9 / 37
Problems Continuous Random Variables
Question (a): Is f (x) a density function?
To determine whether a function f (x) is a p.d.f., we just need to check
_
f
X
(x) 0 x,

f
X
(x)dx = 1.
First, since > 1, we have
f (x) =
_
( 1)x

0 x 1
0 x < 1
.
Second, we have

f (x)dx =

1
( 1)x

dx
= x
+1

1
= 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 10 / 37
Problems Continuous Random Variables
Question (a): Is f (x) a density function?
To determine whether a function f (x) is a p.d.f., we just need to check
_
f
X
(x) 0 x,

f
X
(x)dx = 1.
First, since > 1, we have
f (x) =
_
( 1)x

0 x 1
0 x < 1
.
Second, we have

f (x)dx =

1
( 1)x

dx
= x
+1

1
= 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 10 / 37
Problems Continuous Random Variables
Question (a): Is f (x) a density function?
To determine whether a function f (x) is a p.d.f., we just need to check
_
f
X
(x) 0 x,

f
X
(x)dx = 1.
First, since > 1, we have
f (x) =
_
( 1)x

0 x 1
0 x < 1
.
Second, we have

f (x)dx =

1
( 1)x

dx
= x
+1

1
= 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 10 / 37
Problems Continuous Random Variables
Question (b): Find F(x)
By denition, we have for any x 1
F(x) =

f (y)dy
=

x
1
( 1)y

dy
= y
+1

x
1
= 1 x
1
.
Therefore
F(x) =
_
1 x
1
x 1
0 x < 1
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 11 / 37
Problems Continuous Random Variables
Question (b): Find F(x)
By denition, we have for any x 1
F(x) =

f (y)dy
=

x
1
( 1)y

dy
= y
+1

x
1
= 1 x
1
.
Therefore
F(x) =
_
1 x
1
x 1
0 x < 1
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 11 / 37
Problems Continuous Random Variables
A college professor never nishes his lecture before the bell rings but
always nishes his lecture within 2 min after the bell rings. Let X =time
that elapses between the ring and the end of the lecture. Suppose the
probability density function of X is
f (x) =
_
kx
2
0 x 2
0 otherwise
.
(a) Find the value of k.
(b) What is the probability that the lecture ends within 1 min of the bell
ringing?
(c) What is the probability that the lecture continues for at least 90 sec
beyond the ring?
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 12 / 37
Problems Continuous Random Variables
A college professor never nishes his lecture before the bell rings but
always nishes his lecture within 2 min after the bell rings. Let X =time
that elapses between the ring and the end of the lecture. Suppose the
probability density function of X is
f (x) =
_
kx
2
0 x 2
0 otherwise
.
(a) Find the value of k.
(b) What is the probability that the lecture ends within 1 min of the bell
ringing?
(c) What is the probability that the lecture continues for at least 90 sec
beyond the ring?
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 12 / 37
Problems Continuous Random Variables
A college professor never nishes his lecture before the bell rings but
always nishes his lecture within 2 min after the bell rings. Let X =time
that elapses between the ring and the end of the lecture. Suppose the
probability density function of X is
f (x) =
_
kx
2
0 x 2
0 otherwise
.
(a) Find the value of k.
(b) What is the probability that the lecture ends within 1 min of the bell
ringing?
(c) What is the probability that the lecture continues for at least 90 sec
beyond the ring?
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 12 / 37
Problems Continuous Random Variables
A college professor never nishes his lecture before the bell rings but
always nishes his lecture within 2 min after the bell rings. Let X =time
that elapses between the ring and the end of the lecture. Suppose the
probability density function of X is
f (x) =
_
kx
2
0 x 2
0 otherwise
.
(a) Find the value of k.
(b) What is the probability that the lecture ends within 1 min of the bell
ringing?
(c) What is the probability that the lecture continues for at least 90 sec
beyond the ring?
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 12 / 37
Problems Continuous Random Variables
Question (a): Find the value of k
By using the property of the p.d.f. we have
1 =

f (x)dx
=

2
0
kx
2
dx
=
x
3
3
k

2
0
=
8
3
k.
It implies that
k =
3
8
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 13 / 37
Problems Continuous Random Variables
Question (a): Find the value of k
By using the property of the p.d.f. we have
1 =

f (x)dx
=

2
0
kx
2
dx
=
x
3
3
k

2
0
=
8
3
k.
It implies that
k =
3
8
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 13 / 37
Problems Continuous Random Variables
Question (b&c): End within 1 min? Continue for 90 sec?
For Q (b), the probability can be calculated as
P(X 1) =

1
0
3
8
x
2
dx =
x
3
3

3
8

1
0
=
1
8
.
For Q (c), the probability can be calculated as
P(X 1.5) =

2
1.5
3
8
x
2
dx
=
x
3
3

3
8

2
1.5
=
3
8
_
8
3

1.5
3
3
_
= 0.578.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 14 / 37
Problems Continuous Random Variables
Question (b&c): End within 1 min? Continue for 90 sec?
For Q (b), the probability can be calculated as
P(X 1) =

1
0
3
8
x
2
dx =
x
3
3

3
8

1
0
=
1
8
.
For Q (c), the probability can be calculated as
P(X 1.5) =

2
1.5
3
8
x
2
dx
=
x
3
3

3
8

2
1.5
=
3
8
_
8
3

1.5
3
3
_
= 0.578.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 14 / 37
Problems Continuous Random Variables
Independent observations are made on a random variable X whose
probability density function is
f (x) =
_
0.125x 0 x 4
0 otherwise
.
(a) Find the mean and variance of X.
(b) Find P(X > 3).
(c) Find the probability that at least three observations are required to
obtain a result of x > 3.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 15 / 37
Problems Continuous Random Variables
Independent observations are made on a random variable X whose
probability density function is
f (x) =
_
0.125x 0 x 4
0 otherwise
.
(a) Find the mean and variance of X.
(b) Find P(X > 3).
(c) Find the probability that at least three observations are required to
obtain a result of x > 3.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 15 / 37
Problems Continuous Random Variables
Independent observations are made on a random variable X whose
probability density function is
f (x) =
_
0.125x 0 x 4
0 otherwise
.
(a) Find the mean and variance of X.
(b) Find P(X > 3).
(c) Find the probability that at least three observations are required to
obtain a result of x > 3.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 15 / 37
Problems Continuous Random Variables
Independent observations are made on a random variable X whose
probability density function is
f (x) =
_
0.125x 0 x 4
0 otherwise
.
(a) Find the mean and variance of X.
(b) Find P(X > 3).
(c) Find the probability that at least three observations are required to
obtain a result of x > 3.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 15 / 37
Problems Continuous Random Variables
Question (a): Find the mean and variance of X
E(X) =

x f (x)dx =

4
0
0.125x
2
dx = 0.125
x
3
3

4
0
=
8
3
.
Var (X) = E
_
(X EX)
2
_
= E(X
2
) (EX)
2
=

x
2
f (x)dx (
8
3
)
2
=

4
0
0.125x
3
dx (
8
3
)
2
= 0.125
x
4
4

4
0

64
9
=
8
9
.
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 16 / 37
Problems Continuous Random Variables
Question (a): Find the mean and variance of X
E(X) =

x f (x)dx =

4
0
0.125x
2
dx = 0.125
x
3
3

4
0
=
8
3
.
Var (X) = E
_
(X EX)
2
_
= E(X
2
) (EX)
2
=

x
2
f (x)dx (
8
3
)
2
=

4
0
0.125x
3
dx (
8
3
)
2
= 0.125
x
4
4

4
0

64
9
=
8
9
.
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 16 / 37
Problems Continuous Random Variables
Question(b): Find P(X > 3)
The P(X > 3) can be calculated as follows
P(X > 3) =

4
3
f (x)dx
=

4
3
0.125 xdx
= 0.125
x
2
2

4
3
=
7
16
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 17 / 37
Problems Continuous Random Variables
Question(c): At least three observations are required
Let Y =the number of observations required to obtain a result of x > 3,
then
Y Geometric(
7
16
).
We just need to calculate
P(Y 3) = (1
7
16
)
2
=
81
256
,
since that at least three trialsare required is equivalent with that the
rst two trialsare failure.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 18 / 37
Problems Continuous Random Variables
Question(c): At least three observations are required
Let Y =the number of observations required to obtain a result of x > 3,
then
Y Geometric(
7
16
).
We just need to calculate
P(Y 3) = (1
7
16
)
2
=
81
256
,
since that at least three trialsare required is equivalent with that the
rst two trialsare failure.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 18 / 37
Problems Some Questions from Assignments
Game Plan
1
Review of Key Concepts
Basic Denitions
Comparison Between Discrete and Continuous Random Variables
2
Problems
Continuous Random Variables
Some Questions from Assignments
3
Ending
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 19 / 37
Problems Some Questions from Assignments
Assume that every time you buy an item of the Hong Kong Disney series,
you receive one of the four types of cards, each with a cartoon character
Mickey, Minnie, Donald and Daisy with an equal probability. Over a period
of time, you buy 6 items of the series. What is the probability that you will
get all four cards?
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 20 / 37
Problems Some Questions from Assignments
Method (1): Three of a kind(e.g. A 3, B, C, D)
The probability that the six items show A, A, A, B, C, D in that order is:
(
1
4
)
6
The no. of ways of arranging A, A, A, B, C, D in a sequence is
6!
3! 1! 1! 1!
The no. of ways of determining which letter appears three times is
4
C
1
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 21 / 37
Problems Some Questions from Assignments
Method (1): Three of a kind(e.g. A 3, B, C, D)
The probability that the six items show A, A, A, B, C, D in that order is:
(
1
4
)
6
The no. of ways of arranging A, A, A, B, C, D in a sequence is
6!
3! 1! 1! 1!
The no. of ways of determining which letter appears three times is
4
C
1
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 21 / 37
Problems Some Questions from Assignments
Method (1): Three of a kind(e.g. A 3, B, C, D)
The probability that the six items show A, A, A, B, C, D in that order is:
(
1
4
)
6
The no. of ways of arranging A, A, A, B, C, D in a sequence is
6!
3! 1! 1! 1!
The no. of ways of determining which letter appears three times is
4
C
1
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 21 / 37
Problems Some Questions from Assignments
Method (1): Two pairs(e.g. A 2, B 2, C, D)
The probability that the six items show A, A, B, B, C, D in that order is:
(
1
4
)
6
The no. of ways of arranging A, A, B, B, C, D in a sequence is
6!
2! 2! 1! 1!
The no. of ways of determining which letters appear twice is
4
C
2
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 22 / 37
Problems Some Questions from Assignments
Method (1): Two pairs(e.g. A 2, B 2, C, D)
The probability that the six items show A, A, B, B, C, D in that order is:
(
1
4
)
6
The no. of ways of arranging A, A, B, B, C, D in a sequence is
6!
2! 2! 1! 1!
The no. of ways of determining which letters appear twice is
4
C
2
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 22 / 37
Problems Some Questions from Assignments
Method (1): Two pairs(e.g. A 2, B 2, C, D)
The probability that the six items show A, A, B, B, C, D in that order is:
(
1
4
)
6
The no. of ways of arranging A, A, B, B, C, D in a sequence is
6!
2! 2! 1! 1!
The no. of ways of determining which letters appear twice is
4
C
2
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 22 / 37
Problems Some Questions from Assignments
Method (1): Final result
In sum, the probability that we will collect all of the four letters is
(
1
4
)
6

6!
3! 1! 1! 1!

4
C
1
+ (
1
4
)
6

6!
2! 2! 1! 1!

4
C
2
.
The nal answer is
1560
4
6
= 0.3809.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 23 / 37
Problems Some Questions from Assignments
Method (1): Final result
In sum, the probability that we will collect all of the four letters is
(
1
4
)
6

6!
3! 1! 1! 1!

4
C
1
+ (
1
4
)
6

6!
2! 2! 1! 1!

4
C
2
.
The nal answer is
1560
4
6
= 0.3809.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 23 / 37
Problems Some Questions from Assignments
Method (2): Consider whether the letter is a dierent one
Let N denote that the letter is a new one;
Let O denote that the letter is an old one.
For example, the probability for the result of
NNONN
can be calculated
1
3
4

2
4

2
4

1
4
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 24 / 37
Problems Some Questions from Assignments
Method (2): Consider whether the letter is a dierent one
Let N denote that the letter is a new one;
Let O denote that the letter is an old one.
For example, the probability for the result of
NNONN
can be calculated
1
3
4

2
4

2
4

1
4
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 24 / 37
Problems Some Questions from Assignments
Method (2) (Cont.)
Number Outcome Probability
4 NNNN 1
3
4

2
4

1
4
NONNN 1
1
4

3
4

2
4

1
4
5 NNONN 1
3
4

2
4

2
4

1
4
NNNON 1
3
4

2
4

3
4

1
4
NNNOON 1
3
4

2
4

3
4

3
4

1
4
NOONNN 1
1
4

1
4

3
4

2
4

1
4
6 NNOONN 1
3
4

2
4

2
4

2
4

1
4
NONNON 1
1
4

3
4

2
4

3
4

1
4
NONONN 1
1
4

3
4

2
4

2
4

1
4
NNONON 1
3
4

2
4

2
4

3
4

1
4
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 25 / 37
Problems Some Questions from Assignments
If n persons, including A and B, are randomly arranged in a straight line,
what is the probability that there are exactly r persons in the line between
A and B.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 26 / 37
Problems Some Questions from Assignments
Method(1): Select two positions from n positions
It is essentially to select two positions from the n positions in a line.
First, the no. of ways to select two positions from the n positions such
that there are exactly r positions between them is
n r 1.
Second, the no. of ways to select two positions from the n positions is
n
C
2
.
Therefore, the probability required to calculate is
n r 1
n
C
2
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 27 / 37
Problems Some Questions from Assignments
Method(1): Select two positions from n positions
It is essentially to select two positions from the n positions in a line.
First, the no. of ways to select two positions from the n positions such
that there are exactly r positions between them is
n r 1.
Second, the no. of ways to select two positions from the n positions is
n
C
2
.
Therefore, the probability required to calculate is
n r 1
n
C
2
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 27 / 37
Problems Some Questions from Assignments
Method(1): Select two positions from n positions
It is essentially to select two positions from the n positions in a line.
First, the no. of ways to select two positions from the n positions such
that there are exactly r positions between them is
n r 1.
Second, the no. of ways to select two positions from the n positions is
n
C
2
.
Therefore, the probability required to calculate is
n r 1
n
C
2
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 27 / 37
Problems Some Questions from Assignments
Method(1): Select two positions from n positions
It is essentially to select two positions from the n positions in a line.
First, the no. of ways to select two positions from the n positions such
that there are exactly r positions between them is
n r 1.
Second, the no. of ways to select two positions from the n positions is
n
C
2
.
Therefore, the probability required to calculate is
n r 1
n
C
2
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 27 / 37
Problems Some Questions from Assignments
Method(2): No. of solutions to an equation
Let L =no. of persons on the left of the two persons;
Let M =no. of persons between the two persons;
Let R =no. of persons on the right of the two persons. Then we have
L + M + R = n 2.
First, if we require M = r , then the equation can be deduced to
L + R = n 2 r .
The no. of solutions to this equation is
nr 1
C
1
= n r 1.
Second, the no. of solutions to the original equation is
n
C
2
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 28 / 37
Problems Some Questions from Assignments
Method(2): No. of solutions to an equation
Let L =no. of persons on the left of the two persons;
Let M =no. of persons between the two persons;
Let R =no. of persons on the right of the two persons. Then we have
L + M + R = n 2.
First, if we require M = r , then the equation can be deduced to
L + R = n 2 r .
The no. of solutions to this equation is
nr 1
C
1
= n r 1.
Second, the no. of solutions to the original equation is
n
C
2
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 28 / 37
Problems Some Questions from Assignments
Method(2): No. of solutions to an equation
Let L =no. of persons on the left of the two persons;
Let M =no. of persons between the two persons;
Let R =no. of persons on the right of the two persons. Then we have
L + M + R = n 2.
First, if we require M = r , then the equation can be deduced to
L + R = n 2 r .
The no. of solutions to this equation is
nr 1
C
1
= n r 1.
Second, the no. of solutions to the original equation is
n
C
2
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 28 / 37
Problems Some Questions from Assignments
Method(2): No. of solutions to an equation
Let L =no. of persons on the left of the two persons;
Let M =no. of persons between the two persons;
Let R =no. of persons on the right of the two persons. Then we have
L + M + R = n 2.
First, if we require M = r , then the equation can be deduced to
L + R = n 2 r .
The no. of solutions to this equation is
nr 1
C
1
= n r 1.
Second, the no. of solutions to the original equation is
n
C
2
.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 28 / 37
Problems Some Questions from Assignments
A communication system consists of n components each of which will,
independently, function with probability p. The whole system will be able
to operate eectively if at least one half-of its components function. In
general, when is a (2k + 1) component system better than a (2k 1)
component system?
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 29 / 37
Problems Some Questions from Assignments
The (2k + 1) component system
We divide it into two parts:
(2k 1) component sub-system + (2) component sub-system.
Let X
1
=No. of components which can function in the (2k 1) - system
X
1
Bin(2k 1, p).
Let Y =No. of components which can function in the (2) - system
Y Bin(2, p).
To make this (2k + 1) component system operate, we need to have
X
1
+ Y k + 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 30 / 37
Problems Some Questions from Assignments
The (2k + 1) component system
We divide it into two parts:
(2k 1) component sub-system + (2) component sub-system.
Let X
1
=No. of components which can function in the (2k 1) - system
X
1
Bin(2k 1, p).
Let Y =No. of components which can function in the (2) - system
Y Bin(2, p).
To make this (2k + 1) component system operate, we need to have
X
1
+ Y k + 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 30 / 37
Problems Some Questions from Assignments
The (2k + 1) component system
We divide it into two parts:
(2k 1) component sub-system + (2) component sub-system.
Let X
1
=No. of components which can function in the (2k 1) - system
X
1
Bin(2k 1, p).
Let Y =No. of components which can function in the (2) - system
Y Bin(2, p).
To make this (2k + 1) component system operate, we need to have
X
1
+ Y k + 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 30 / 37
Problems Some Questions from Assignments
The (2k + 1) component system
We divide it into two parts:
(2k 1) component sub-system + (2) component sub-system.
Let X
1
=No. of components which can function in the (2k 1) - system
X
1
Bin(2k 1, p).
Let Y =No. of components which can function in the (2) - system
Y Bin(2, p).
To make this (2k + 1) component system operate, we need to have
X
1
+ Y k + 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 30 / 37
Problems Some Questions from Assignments
The (2k 1) component system
Let X
2
=No. of components which can function in this system, and
X
2
Bin(2k 1, p).
To make this (2k 1) component system operate, we need to have
X
2
k.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 31 / 37
Problems Some Questions from Assignments
The (2k 1) component system
Let X
2
=No. of components which can function in this system, and
X
2
Bin(2k 1, p).
To make this (2k 1) component system operate, we need to have
X
2
k.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 31 / 37
Problems Some Questions from Assignments
(2k + 1) - system (X
1
+Y) v.s. (2k 1) - system (X
2
)
For the (2k + 1) component system, we calculate
P(X
1
+ Y k + 1)
= P(X
1
k 1 Y = 2) + P(X
1
k Y = 1) + P(X
1
k + 1 Y = 0).
Hence,
P(X
1
+ Y k + 1)
= P(X
1
k Y = 2) + P(X
1
= k 1 Y = 2)
+ P(X
1
k Y = 1)
+ P(X
1
k Y = 0) P(X
1
= k Y = 0).
Therefore, the probability that it can operate is
P(X
1
k) + [P(X
1
= k 1) P(Y = 2) P(X
1
= k)P(Y = 0)] .
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 32 / 37
Problems Some Questions from Assignments
(2k + 1) - system (X
1
+Y) v.s. (2k 1) - system (X
2
)
For the (2k + 1) component system, we calculate
P(X
1
+ Y k + 1)
= P(X
1
k 1 Y = 2) + P(X
1
k Y = 1) + P(X
1
k + 1 Y = 0).
Hence,
P(X
1
+ Y k + 1)
= P(X
1
k Y = 2) + P(X
1
= k 1 Y = 2)
+ P(X
1
k Y = 1)
+ P(X
1
k Y = 0) P(X
1
= k Y = 0).
Therefore, the probability that it can operate is
P(X
1
k) + [P(X
1
= k 1) P(Y = 2) P(X
1
= k)P(Y = 0)] .
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 32 / 37
Problems Some Questions from Assignments
(2k + 1) - system (X
1
+Y) v.s. (2k 1) - system (X
2
)
For the (2k + 1) component system, we calculate
P(X
1
+ Y k + 1)
= P(X
1
k 1 Y = 2) + P(X
1
k Y = 1) + P(X
1
k + 1 Y = 0).
Hence,
P(X
1
+ Y k + 1)
= P(X
1
k Y = 2) + P(X
1
= k 1 Y = 2)
+ P(X
1
k Y = 1)
+ P(X
1
k Y = 0) P(X
1
= k Y = 0).
Therefore, the probability that it can operate is
P(X
1
k) + [P(X
1
= k 1) P(Y = 2) P(X
1
= k)P(Y = 0)] .
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 32 / 37
Problems Some Questions from Assignments
Final result
To make the (2k + 1) - system better than the (2k 1) - system, we just
need to make
P(X
1
= k 1) P(Y = 2) > P(X
1
= k)P(Y = 0).
This is equivalent to
2k1
C
k1
p
k1
(1 p)
k
p
2

2k1
C
k
p
k
(1 p)
k1
(1 p)
2
.
Hence,
p 1 p
1
2
p 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 33 / 37
Problems Some Questions from Assignments
Final result
To make the (2k + 1) - system better than the (2k 1) - system, we just
need to make
P(X
1
= k 1) P(Y = 2) > P(X
1
= k)P(Y = 0).
This is equivalent to
2k1
C
k1
p
k1
(1 p)
k
p
2

2k1
C
k
p
k
(1 p)
k1
(1 p)
2
.
Hence,
p 1 p
1
2
p 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 33 / 37
Problems Some Questions from Assignments
Final result
To make the (2k + 1) - system better than the (2k 1) - system, we just
need to make
P(X
1
= k 1) P(Y = 2) > P(X
1
= k)P(Y = 0).
This is equivalent to
2k1
C
k1
p
k1
(1 p)
k
p
2

2k1
C
k
p
k
(1 p)
k1
(1 p)
2
.
Hence,
p 1 p
1
2
p 1.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 33 / 37
Problems Some Questions from Assignments
Exactly one of six similar keys opens a certain door. If you try the keys one
after another, what is the expected number of keys that you will have to
try before success?
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 34 / 37
Problems Some Questions from Assignments
Solution
Note: The number of keys required to succeed does not follow a geometric
distribution, since the rst trial is not independent with the second trial.
To try the keys one after another is essentially to arrange six keys in a
sequence.
The correct key is equally likely to be one of the six ones and the number
of trials required follows a uniform distribution with possible values:
1, 2, 3, 4, 5, 6.
Its expectation can be given by
1 + 2 + 3 + 4 + 5 + 6
6
= 3.5.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 35 / 37
Problems Some Questions from Assignments
Solution
Note: The number of keys required to succeed does not follow a geometric
distribution, since the rst trial is not independent with the second trial.
To try the keys one after another is essentially to arrange six keys in a
sequence.
The correct key is equally likely to be one of the six ones and the number
of trials required follows a uniform distribution with possible values:
1, 2, 3, 4, 5, 6.
Its expectation can be given by
1 + 2 + 3 + 4 + 5 + 6
6
= 3.5.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 35 / 37
Problems Some Questions from Assignments
Solution
Note: The number of keys required to succeed does not follow a geometric
distribution, since the rst trial is not independent with the second trial.
To try the keys one after another is essentially to arrange six keys in a
sequence.
The correct key is equally likely to be one of the six ones and the number
of trials required follows a uniform distribution with possible values:
1, 2, 3, 4, 5, 6.
Its expectation can be given by
1 + 2 + 3 + 4 + 5 + 6
6
= 3.5.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 35 / 37
Problems Some Questions from Assignments
Solution
Note: The number of keys required to succeed does not follow a geometric
distribution, since the rst trial is not independent with the second trial.
To try the keys one after another is essentially to arrange six keys in a
sequence.
The correct key is equally likely to be one of the six ones and the number
of trials required follows a uniform distribution with possible values:
1, 2, 3, 4, 5, 6.
Its expectation can be given by
1 + 2 + 3 + 4 + 5 + 6
6
= 3.5.
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 35 / 37
Ending
Game Plan
1
Review of Key Concepts
Basic Denitions
Comparison Between Discrete and Continuous Random Variables
2
Problems
Continuous Random Variables
Some Questions from Assignments
3
Ending
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 36 / 37
Ending
Goodbye and See You
See You After Reading Week (Oct 24,26)
Same Time
Same Place
Jun FU (S&AS, HKU) STAT1306 Introductory Statistics. October 10, 2011 37 / 37