Академический Документы
Профессиональный Документы
Культура Документы
SET THEORY
A. General Learning Objectives
Understanding basic concepts of set theory and able to apply it in sigma field
concept and function set.
B. Specific Learning Objectives
After learning this chapter, you are expected able to:
1. Determining whether a given set is a field or not.
2. Determining whether a given set is a-algebra or not.
3. Determining the value of a function set
1.1 Some Definition and Notation
We already know the general term of the set (S), that is a (well dened) collection
of distinct objects which we denote by s. The fact that s is a member of S, an
element of S, or that it belongs to S is expressed by writing s S. The negation
of the statement is expressed by writing s S. We say that S is a subset of S,
or that S is contained in S, and write S S, if for every s S , we have s S.
S is said to be a proper subset of S, and we write S S, if S S and there
exists s S such that s S . Sets are denoted by capital letters, while lower case
letters are used for elements of sets.
Definition 1.1
Power set of the set X is the set of all subsets of X. Power set usually denoted by
P(X) or 2 X
Example:
If X={a,b} then P( X ) {,{a},{b}, X }
In set theory we have to know some operations involving two or more sets, for
example:
1. The complement (with respect to S ) of the set A, denoted by Ac ,
is dened by Ac = {s S ; s A}
2. The union of the sets Aj , j = 1, 2, . . . , n, to be denoted by A 1 A 2
A3
An
A j = { x : x
j 1
or A j difined by
j 1
A3 An
A
j 1
or A
j 1
difined by
= { x : x A j foe every j }
1. A
B=B
B=B
A
A
( B C ) =
A ( B C ) =
2. A
(commutative)
(A B )
C
( A B ) C
(Associative)
( B C ) = (A B ) (A C )
A ( B C ) = (A B ) (A C )
3. A
De Morgan Law
(Distributive)
i.
A j
(
=
j
A j )c
ii.
A j
(A j )c
j
A j
j
A j )c
(
j
(
and
j
A j )c
A j
j
We will then, by denition, have veried the desired equality of the two
sets.
c
A j
a. let s
. Thus, s
A j
(
other word, s A cj for every j. therefore, s
j
(
b. Let s
j
A j )c
, then s
A cj
A j )c
j. Then
A j
. and therefore s
A j
j
defined by a single point but is determined by the entire set of points. Such
function is called set function
1.4 Field and
-Field
Definisi 1.2
A class (set) of subsets of
a.
F is non-empty class
b.
if A F then Ac F
c.
if A1 , A2 , , An
A1 , A2 ,
F then
then
i 1
i 1
A i
A i
F. if ( c ) replace by
F then
F is called by
(c ) that is, if
-field
example:
let A is subset of S, and let F = { , , A ,
-field
therefore F is
Definition 1.3
let measurable space (S, F F). A measurable ( . )is set function that difined in F
satisfied
(i ) (A) 0 , A F .
(ii)
i 1
A i
i 1
A i
Chapter II
PROBABILITY
A. General Learning Objectives
Understanding the concept of probability and able to applied those concepts
to calculate the probability of an event.
B. Specific Learning Objectives
After learning this chapter, you are expected able to:
4. Calculate probabiity of events with or without replacement
5. Calculate conditional probability of the given event
2.1 Introduction
Historically, the oldest way of measuring probabilities, the clasical probability
concept, applies when all possible outcomes are equally likely, as is presumbly the
case in most games of chance. We can then say that if there are N equally likely
possibilities, of which one must occur and n are regarded as favorable, or as a
success then the probability of a succes is given by the ratio Nn .
Example:
What is probability of rolling a 3 with a die?
Solution;
1
6
Although equally likely possib ilities are found mostly in games of chance, the
classical probability concept applies also in a great variety of situations where
gambling devices are used to make random selections...
2.2 Sample Space
In probability theory, the sample space of an experiment or random trial is
the set of all possible outcomes or results of that experiment. A sample space is
usually denoted using set notation, and the possible outcomes are listed
Postulate 1. The probability of an event is a nonnegative real number; that is P( A) 0 for any
subset A of S
Postulate 2. P( S ) 1
Postulate 3. If A1 , A2 ,... is a finite or infinite
sequence of mutually exclusive event of S, then
P ( A1 A2 ...) P ( A1 ) P ( A2 ) ...
Postulate above doesnt need to proof but if the resulting theory is to be applied,
we must show that the postulates are satisfied when we give probabilities a real
meaning.
Example:
For each of the following, explain why it is not a permissible way of assigning
probabilities to the four possible and mutually exclusive outcomes A,B,C and D
a. P ( A) 0.12, P ( B ) 0.55, P (C ) 0.2, P( D) 0.1
b. P( A) 0.12, P ( B ) 0.55, P (C ) 0.01, P ( D ) 0.1
Solution:
In (a) we find that P ( D) 0.1 violates postulate 1, and in (b) we get
P ( S ) P( A B C D ) 0.12 0.55 0.01 0.1 0.78 which violates postulate
2.
Theorem 2.1
If A is an event in a discrete sample space S, then P(A) equals the sum of the
probabilities of the individual outcomes comprising A.
Proof:
Let M 1 , M 2 , M 3 ,... be the finite or infinite sequence of outcomes which comprise
the event A, Thus
A M 1 M 2 M 3 ...
And since the individual outcomes, the Ms are by definition mutually exclusive,
the third postulate of probability yields
P ( A) P ( M 1 ) P( M 2 ) P( M 3 ) ...
This complete the proof.
Example
If a balance coin is tossed twice, what is the probability of getting at least one
head?
Solution
The sample space for this experiment is
S {HH , HT , TH , TT }
Since the coin is balanced, we assume that each of these outcomes is equally
likely to occur, and we therefore assign a probability of
1
4
A is the event that we will get at least one head, then A{HH , HT , TH } and
P( A) P( HH ) P ( HT ) P (TH )
14 14 14
3
4
Theorem 2.2
If experiment can result in any one of N different equally likely outcomes, and if n
of these outcomes together constitute event A, then the probability of event A is
P( A)
n
N
Proof:
Let M 1 , M 2 , M 3 ,...M N represent the individual outcomes is S, each with
probability
1
N
1 1 1
1
...
N4 4
N 4 2N4 4 43N
1
nterms
P( A)
n
N
Thus, P( A' ) 1 P ( A)
Theorem 2.4
P () 0 for any sample space S
Proof:
Since S S and the events S and are mutually exclusive , it follows that
P( S ) P( S )
P ( S ) P ()
Proof:
Using theorem 2.5 and the fact that A S for any event A in S , we have
Proof:
Assigning the probabilities a,b,and c to the mutually exclusive events
A B, A B ' , and A' B as in the venn diagram of figure 2.5, we find that
P( A B) a b c
( a b ) (c a ) a
P ( A) P ( B) P ( A B )
B
b
Theorem 2.8
If A,B, and C are any three events in a sample space S, then
P ( A B C ) P ( A) P ( B ) P (C ) P ( A B ) P ( A C ) P ( B C ) P ( A B C )
proof:
Writing A B C as A ( B C ) and using therem 2.7 twice , once for
P ( A ( B C )) and once for P ( B C ) , we get
P ( A B C ) P ( A ( B C ))
P ( A) P( B C ) P ( A ( B C ))
P ( A) P( B) P (C ) P ( B C ) P ( A ( B C ))
P ( A B C ) P ( A) P ( B ) P (C ) P ( A B ) P ( A C ) P ( B C ) P ( A B C )
X 6 ,
distribution function m( )
X 4 .
We assign the
1
1
for 1, 2,3,..., 6 . Thus, P( F ) . Now suppose
6
6
that the die is rolled and we are told that the event E has occurred. This leaves
only two possible outcomes: 5 and 6. In the absence of any other information, we
would still regard these outcomes to be equally likely, so the probability of F
becomes
1
1
, making P( F | E ) .
2
2
Example:
A consumer research organization has studied the services under warranty
provided by the 50 new car dealers in a certain city, and its findings are
summarized in the following table:
Good service
Poor service
Under warranty
In business ten years or 16
Under warranty
4
more
In business less than ten 10
20
years
If a person randomly selects one of these new car dealers, what is the probability
that he gets one who provides good service under warranty? Also, if a person
randomly selects one of the dealer who has been in business for ten years or more
what is the probability that he gets one who provides good service under
warranty?
Solution:
By randomly we mean that, in each case , all possible selection are equally likely,
and we can therefore use the formula of theorem 2.2. if we let G denote the
selection of the dealer who provides good service under warranty, and if we let
n(G) denote the number of elements in G and n(S) the number of elements in the
whole sample space, we get
P (G )
n(G ) 16 10
0.52
n( S )
50
16
0.8
20
Where T denotes the selection of a dealer who has been in business ten years or
more. This answer the second question, and as should have been expected,
P(G | T ) is considerably higher than P(G).
Since the numerator of P(G | T ) is n(T G ) 16 the number of dealers who have
been in business for ten years or more and provide good service under warranty,
and the denominator is n(T) the number of dealer who have been in business ten
years or more, we can write symbolically
P(G | T )
n(T G )
n(T )
P (G | T )
n ( T G )
n(S )
n (T )
n(S )
P (T G )
P(T )
And we have, thus, expressed the conditional probability P(G | T ) in terms of two
probabilities defined for the whole sample space S.
Definition 2.1
If A and B are any two events in a sample space S and P( A) 0 , the conditional
probability of B given A is
P ( B | A)
P( A B)
P ( A)
Example:
With reference to previous example, what is the probability that one of the dealers
who has been in business less than ten years will provide good service under
warranty?
Solution:
'
Since P (T G )
10
10 20
0.2 and P (T ' )
0.6 substitution into the formula
50
50
yields
P(G | T ' )
P(T ' )
0.6 3
Theorem 2.9
If A and B are any two events in a sample space S and P( A) 0 , then
P ( A B ) P ( A) P ( B | A)
P ( A) P ( Bi ) P ( A | Bi )
i 1
P( Br ) P( A | Br )
k
P( B ) P( A | B )
i 1
For r 1, 2,..., k
Exercise 2
1. A gum-ball machine gives out a red, a black, or a green gum ball
a. Describe an approrpite sample space
b. List all possible events
c. If R is the event red then list the outcomes in R
d. if G is the event green then what is R G
2. we obtain 100 gum balls from a machine, and we get 20 red (R), 30 black
gum balls (B) and 50 green (G) gum balls.
a. Can we use , as probability model for the color of a gum ball from the
machine,
one
given
by
p3 P(G ) 0.5 ?
b. Suppose we later notice that some yellow (Y) gum balls are also in the
machine . could we use as a model p1 0.2, p2 0.3, p3 0.5 and
p4 0.1 ?
3. A bag contains five blue balls and three red balls. A boy draws a ball, and
then draws another without replacement. Compute the following
probabilities:s
a. P(2 blueballs )
b. P(1blue and 1 red )
c. P (at least blue)
4. Suppose that in Singaraja , the probability that rainy fall day is followed
by a rainy day is 0.8 and the probability that a sunny fall day is followed
by a rainy day is 0.6. find the probabilities that a rainy day is followed by
a. a rainy day, a sunny day, and another rainny day
b. two sunny days and then a rainy day
c. two rainy days and then two sunny days
d. rain two days later.
5. A balanced die tossed twice. If A is the event that an even number comes
up on the first tossed, B is the event that an event number comes up on the
second toss, and C is the event that both tosses result in the same number,
are the event A,B,and C independent?
6.
CHAPTER III
RANDOM VARIABLES AND ITS DISTRIBUTION
A. General Learning Objectives
Construct a mathematical model to describe probability on event in a sample
space.
B. Specific Learning Objectives
After learning this chapter, you are expected able to:
1. Determine the density function of a random variable.
2. Determine the distribution function of a random variable.
3. Calculate the expected value of a random variable.
4. Using the properties of the expected value of random variables to
calculate the expected value of a random variable.
5. Determine the moment generating function of a random variable.
There are two kinds of random variable, that are discrete random variable (defined
over discrete sample space) whose range is finite or countably infinite
and
f ( x) 1 , where the summation extends over all the values within its
x
domain
We left the proof to the reader as exercise
Definition 3.3
If
is
discrete
random
variable,
the
function
given
by
F ( x) P ( X x) f (t ) for x
tx
1. F () 0
2. F () 1
3. If a b , then F (a) F (b) for any real numbur a and b
Theorem 3.3
If the range of a random variable X consists of the values x1 x2 x3 ... xn
then f ( x1 ) F ( x1 ) and
f ( xi ) F ( xi ) F ( xi 1 ),
for i 2,3,..., n
Definition 3.4
If X is discrete random variable
3.3 Continuous Random Variable
Definition 3.5
A function with values f(x), defined over the set of all real numbers, is called a
probability density function of the continuous random variable X if and only if
b
P(a X b) f ( x )dx
a
f ( x ) 0 for x
2.
Example:
f ( x) 1
ke3 x
0
f ( x)
for x 0
elsewhere
f ( x )dx ke 3 x dx
0
e 3 x
k lim
t 3
k
1
3
P (0.5 X 1)
3e
0.5
3 x
dx e 3 x
1
0.5
0.173
Definition 3.6
If X is a continuous random variable, the function given by
F ( x ) P ( X x)
f (t )dt for x
Where f(t) is the value of the probability density function of X at t, is called the
distribution function of comulative distribution of X.
Futheremore, it immediately follows from definition 3.6 that
Theorem 3.5
If f(x) and F(x) are, respectly, values of the probability distribution and the
distribution function of X at x, then
P (a X b) F (b ) F (a )
f ( x)
example.
Also,
use
this
distribution
function
to
reevaluate
P(0.5 X 1)
Solution:
For x 0
F ( x)
f (t )dt 3e 3t dt e 3t 1 e 3 x
0
F ( x)
for x 0
3 x
for x 0
1 e
To find the probability P(0.5 X 1) we make use of the formula of the first part
of theorem 3.5, getting
P (0.5 X 1) F (1) F (0.5)
(1 e 3.1 ) (1 e 3.(0.5) )
0.173
This agrees with the result obtained by using the probability density function
Definition 3.7
If X is discrete random variable and f(x) is the value of its probability distribution
at x, the expected value of this random variable is
E ( X ) xf ( x)
x
xf ( x)dx
3(1 x) 2
0
for 0 x 1
eslewhere
Let f ( x )
E ( X ) x(1 x ) 2 dx
0
5
4
Proof:
1. E ( a)
Proof properties 2
af ( x )dx a f ( x )dx a
E (aX b)
(ax b) f ( x)dx
(axf ( x) bf ( x))dx
a xf ( x)dx b f ( x)dx
aE ( X ) b
Proof of properties 3 left to the readers
Definition 3.8
The variance of random variable X difined by
Var ( X ) 2 E ( X ) 2 E X 2 2
Dengan E ( X )
Theorem 3.6
If X has the variance 2 , then
Var (aX b) a 2 2
Proof
Var (aX b) E
aX b E aX b
2
E aX b a b
E a 2 ( X )2
a 2 E ( X )2
a
2
Definition 3.9
'
The nthmoment about the origin of the random variable X, denoted by n , is the
n' E ( X n ) x n f ( x)
x
n' E ( X n )
f ( x )dx
When X is continuous.
From definition above we can conclude that,
1. First moment obout the originis mean and notated by
2. First moment about the mean is 0
3. Second moment about mean is variance
Chebyshevs Theorem
Theorem 3.7
If and are, respectively, the mean and the standard deviation of the random
variable X, then for any positive constant k the probability is at least 1
1
that X
k2
1
k2
Proof:
2 E ( X )2
E (X )
2
(x )
f ( x)dx
( x ) f ( x)dx
2
(x )
k
f ( x)dx
(x )
f ( x) dx
( x ) 2 f ( x)dx
(x )
f ( x)dx
k 2 2 f ( x) dx
k
2
f ( x)dx
k2
f ( x)dx
f ( x )dx
Provided 2 0 . Since the sum of the two integrals in this inequality represents
the probability that X will take on a value less than or equal to k or greater
than or equal to k , we have, thus, shown that
P | X | k
1
k2
3.3
1
k2
Although
tx
f ( x )dx
When X is continuous
To explain why we refer to this function as amoment generating function , let us
substitute for etx its Maclaurins series expansions, namely,
etx 1 tx
t 2 x 2 t 3 x3
t n xn
...
...
2!
3!
n!
t 2 x 2 t 3 x3
t n xn
M x (t ) 1 tx
...
... f ( x )
2!
3!
n!
x
2
t
tn
f ( x) t f ( x) x x 2 f ( x) ... x n f ( x) ...
2! x
n! x
x
x
1 t 2'
t2
tn
... n' ...
2!
n!
And it can be seen that in the Maclaurins series of the moment generating
function of X the coefficient of
tn
'
is n , the nth moment about the origin of the
n!
e x
0
f ( x)
Solution
By definition
for x 0
elsewhere
M x (t ) E (etX )
tx x
e dx
x (1 t )
dx
1
for | t | 1
1 t
Theorem 3.8
d n M x (t )
n'
dt n t 0
Theorem 3.9
If a and b constant , then
( X a )t
1. M X a (t ) E e
eat M X (t )
bXt
2. M bX (t ) E e M X (bt )
3. M Xba (t ) E e
X a
t
b
a
t
b
t
e MX
1
2. Suppose distribution function F ( x) 1 , for x 0,1, 2,3,...
2
a. Determine density function of X
b. Determine Pr(10 X 20)
c. Determine probability of X where X is even
x
, x 1, 2, 5 ,
8
determine
a. E ( X )
b. Var ( X )
c. E (2 X 3)
4. Let X is random variable with density function f ( x) 3 x 2 , 0 x 1 ,
determine
a. E ( X )
b. Var ( X )
c. E ( X 2 )
d. E (3 X 5 X 2 1)
5. Let X is random variable with density function f ( x) e ( x 2) , for
2 x , determine
a. MGF of X
b. Use the result in (a) to determine E ( X ) and E ( X 2 )
CHAPTER IV
SPECIAL PROBABILITY DISTRIBUTIONS
for x 0,1
Definition 4.2
Example
Find the probability of getting 5 heads and 7 tails in 12 flips of a balanced coin
Solution
Substituting
the
formula
for
the
binomial
distirbution, we get
12 5
12 5
0.19
(1 0.5)
5
b( x; n, )
Theorem 4.1
The mean and the variance of the binomial distribution are
n and 2 n (1 )
Proof:
To determine the mean, let us directly evaluate the sum
n
n
n x
x x 1
x 0 x
n!
n x
x 1
x 1 ( x 1)( n x )!
Where we omitted the term corresponding to x=0 , which is 0 and the canceled
n
the x againts the first factor of x!=x(x-1)! in the denominator on
x
Then, factoring out the factor n in n!=n(n-1)! And one factor , we get
n 1
n x
x 1
1
x 1 x 1
(proved)
'
To find expression for 2 and then 2 , let us make use of the fact that
n!
x (1 )n x
x 2 ( x 2)!( n x )!
E X ( X 1)
n 2
x 2
n x
(1 )
x 2 x 2
n
E X ( X 1) n(n 1) 2
Therefore,
2' E X ( X 1) E ( X ) n(n 1) 2 n
And, finally
2 2' 2
n( n 1) 2 n n 2 2
n (1 )
In such situations we are often interested in whether the events occur randomly in
time or space
Definition 4.3
A random variable X has a poisson distribution and it is reffered to as poisson
random variable , if and only if its probability distribution is given by
p ( x; )
x e
x!
Example:
If the probability is 0.005 that any one person attending a parade on a very hot day
will suffer from heat exhaustion, what is the probability that 18 of the 3000 person
attending the parade will suffer from heat exhaustion?
Solution
Substituting x=18 and 3000(0.005) 15 into the formula for the poissoon
distribution, we get
p (18;15)
1518 e 15
0.0706
18!
Theorem 4.2
The mean and the variance of the poisson distribution are
and 2
Theorem 4.3
The moment generating function of the poisson distribution is given by
t
M X (t ) e ( e 1)
Proof:
x e
( e t ) x
M x (t ) e
e
x!
x!
x 0
x 0
xt
( e t ) x
Where
can be recognized as the maclaurins series of e z with z et .
x
!
x 0
Thus,
t
M x (t ) e e e e ( e 1)
Then, differentiating M X (t ) twice with respect to t, we get
t
M X' (t ) et e ( e 1)
t
M X'' (t ) et e ( e 1) 2e 2t e ( e 1)
So
'
thus
and
2 2' 2 .
4.1.4
Hypergeometric Distribution
x n x
h( x; n, N , k )
N
n
for x 0,1, 2,..., n
x k and n x N k
Thus, for sampling without replacement, the number of success in n trials is a
random variable having a hypergeometric distribution with the parameter n,N,
and k.
Example:
Think an basket wich contains two types of balls, blacks and whites. Define
drawing a white ball as a success and drawing a black ball as a failure (analogous
to the binomial distribution). If the variable N describes the number of all balls in
the basket and k describes the number of white marbles, then N k corresponds to
the
number
of black
balls.
Now, assume (for example) that there are 5 white and 45 black balls in the basket.
Close your eyes and draw 10 balls without replacement.
What is the probability that exactly 4 of the 10 are white?
Solution:
N 50, k 5, n 10, x 4
5 50 5
4 10 4
P ( X 4)
0.00396
50
10
Theorem 4.4
The mean and the variance of the hypergeometric distribution are
nk
nk ( N k )( N n)
and 2
N
N 2 ( N 1)
Proof:
To determine the mean, let us directly evaluate the sum
k N k
n
x n x
x
N
x 0
n
N k
n
n x
k!
N
x 1 ( x 1)!( k x )!
n
Where we ommited the term corresponding to x=0, which is 0, and cancelled the x
k
against the first factor of x!=x(x-1)! In the denominator of , then factoring
x
k
out N , we get
n
k
N
n
k 1 N k
x 1 x 1
n x
n
k
N
n
k 1 N k
y m y
y 0
m
Finally ,
k N 1
k N 1 nk
N
N m N n 1
n
n
E ( X 2 ) E X ( X 1) E ( X )
2 E X ( X 1) E ( X ) 2
E X ( X 1)
k (k 1)n(n 1)
N ( N 1)
We thus obtain
k (k 1)n(n 1) nk nk
N ( N 1)
N N
nk ( N k )( N n)
N 2 ( N 1)
4.1.5
Geometric Distribution
g ( x; p ) p(1 p )i 1 1 (1 p ) x
i 1
1
1 p
and 2 2
p
p
pet
1 (1 p)et
1
n
for x x1 , x2 ,..., xn
Where xi x j , when i j
Theorem 4.6
The mean and the variance of the discrete uniform distribution are
n
1 1
1 (n 2 1)
2
xi (n 1) and xi
n 2
n
12
i 1
i 1
n
1
for x
0
elsewhere
f ( x)
Theorem 4.7
The mean and the variance of the discrete uniform distribution are
( )2
and 2
2
12
Definition 4.8
Gamma function which notated by ( ) of any 0 , difined by
( ) x 1e x dx
0
( 12 )
Definition 4.9
A random variable X has a Gamma distribution and it is reffered to as Gamma
random variable , if and only if its probability distribution is given by
f ( x)
x
1
1
x
e
( )
0
for x 0
elsewhere
Where , 0
To say that any variable random has distribution as above, we use notation
X ~ Gam( , )
M X (t ) E e
tX
tx
M X (t ) E etX
0
x
1
1
x
e
dx
( )
1
(t 1 ) x
x 1e dx
( )
1
Let, u (t ) x , thus
u
du
1
du t dx , and x 1
, dx 1
t
t , therefore,
M X (t )
1
u 1
1
eu 1
du
1
( ) 0 ( t )
( t)
M X (t )
1
1
u 1e u du ( 1 t )
1
( ) ( t ) 0
See that
M X' (t ) (1 t ) 1
M X'' (t ) ( 1) 2 (1 t ) 2
M X''' (t ) ( 1)( 2) 3 (1 t ) 3
.
.
.
M X( n ) (t ) ( 1)( 2)...( n 1) n (1 t ) n
n ( n 1)!(1 t ) n
( 1)!
( n) n (1 t ) n
( )
M X( n ) (0)
( n) n
E(X n )
( )
( 1)
( )
( )
( )
( )
( )
( 1)( ) 2
( 1) 2
( )
E( X 2 )
2 2 2 2 2 2
Definition 4.10
A random variable X has a exponential distribution and it is reffered to as
exponential random variable , if and only if its probability distribution is given by
1 x
e
f ( x)
0
for x 0
elsewhere
Where 0
Notation X ~ Exp( ) means that random variable X has distribution as above. If
in gamma distribution taken 1and , then obtained
f ( x)
1 11 x 1 x
x e e
(1)
Then we can see that exponential distribution is special cases of the gamma
distribution
X ~ Exp ( ) Gam( ,1)
By taking result that obtained in gamma distribution then it is easy to find its
mean and variance
, 2 2 and M X (t ) (1 t ) 1
Another special case of the gamma distribution arises when
v
and 2 .
2
Definition 4.11
A random variable X has a Chi-Square distribution and it is reffered to as ChiSquare random variable , if and only if its probability distribution is given by
v2 x
1
2
x
e 2 for x 0
f ( x) 2 2 ( v2 )
0
elsewhere
v , 2 2v and M X (t ) (1 2t ) v /2
4.2.3
The normal distribution is the most widely known and used of all distributions.
Because the normal distribution approximates many natural phenomena so well, it
has developed into a standard of reference for many probability problems.
Definition 4.12
A random variable X has a Normal distribution and it is reffered to as Normal
random variable , if and only if its probability distribution is given by
n( x; , )
1 x
1
e 2
2
for x
Where 0
Definition 4.13
A normal random variable Z with 0 and 1 is referred to as the standard
normal distribution, with density function ( Z )
Notated Z ~ N (0,1) , and its pdf is
(t )dt
(Z )
1 t22
e dt
2
Theorem 4.8
If X ~ N ( , 2 ) then,
X
~ N (0,1)
X
2. FX ( x) ( Z )
1. Z
Proof 1
X
z
P ( X z )
FZ ( z ) P( Z z ) P
1 x
1
e 2 dx
2
2
1 z22
e
2
x
dx
then dw
Let w
FZ ( z )
1 12 w 2
e
dw ( Z )
2
By Differentiating FZ ( z ) , we obtained
d ( FZ ( z ))
1 z22
fZ ( z)
e dz it means Z ~ N (0,1)
dz
2
Proof 2
FX ( x) P ( X x )
x
x
X x
P Z
Or
x
FX ( x)
Let z
12 w
1
e
dw
2
2
w
dw
x
then dz
, thus for w x , z
Therefore, FX ( x )
1 12 z 2
e
dz ( x )
2
( z ) 1 ( z )
( z )
Excercise 4
t
1. If X is bernoulli random variable , show that M X (t ) q pe
2. If the probability of picking a winning horse in a race is 0.2 , and if X is
a. Pr( X 4)
b. Pr( X 4)
c. E ( X ) and Var ( X )
3. The number of calls that arrive at a switchboard during one hour is poisson
distributed with mean 10 . Find the probability of occurence during an
hour of each of the following events:
a. Exactly seven calls arrive
b. At most seven calls arrive
c. Between three and seven calls (inclusive) arrive
4. The survival time (in days) of a white rat that was subjected to a ceratin
level of X-ray radiation is a random variable X ~ GAM (5, 4)
a. Determine Pr( X 15)
b. Pr(15 X 20)
c. Find the expected survival time , E ( X ) .
5. Assume that the time (in hours) until failure of a transitor is a random
variable X ~ EXP(100)
a. Find the probability that X 15
b. Find the probability that X 110
c. It is observed after 95 hours that the transitor still is working, find the
conditional probability that X 110 . How does this compare to (a)?
Explain this result.
d. What is Var ( X )
6. Suppose that X ~ N (10,16) find:
a. Pr X 14
b. Pr 4 X 18
c. Pr 2 X 10 18
d. X 0.95 , the 95th percentile of X
CHAPTER V
JOINT DISTRIBUTION
variables.
Determine the marginal density function of a random variable if its joint
3.
4.
5.
6.
variable.
Determine the independence of he given trandom variables.
Determine the correlation coefficient of the given random variables.
In many applications there will be more than one random variable of intrest, say
X 1 , X 2 ,..., X k . It is convinient mathematically to regard these variables as
components of a k-dimensional vector, X ( X 1 , X 2 ,..., X k ) , which is capable of
assuming values x ( x1 , x2 ,..., xk ) in a k-dimensional euclidean space. As before
we will develop the discrete and continuous cases separately
5.1 Joint Discrete Distribution
Definition 5.1
The joint probability density function (joint pdf) of the k-dimensional discrete
random variable X ( X 1 , X 2 ,..., X k ) is defined to be
f ( x1 , x2 ,..., xk ) Pr X 1 x1 , X 2 x2 ,..., X k xk
For all possible values x ( x1 , x2 ,..., xk ) of X.
Example:
A box consists of 10 flower seeds. 4 of them is red, 4 white and 2 pink. Suppose 5
seeds taken randomly without replacement , X 1 is number of seeds picked up red
and X 2 is number of seeds picked up white , Then, the joint density function of
X 1 and X 2 defined by
2
4 4
x 1 x 2 5 x 1 x 2
f x1 , x2
, 0 x1 , 0 x2 dan x1 x1 10
10
5
(see hypergeometric distribution)
Probabilitas getting 2 red and 2 white is determined by f 2, 2 .
Definition 5.2
The Joint Cumulative distribution functions of the k random variables
X 1 , X 2 ,..., X k is the function defined by
F ( x1 , x2 ,..., xk ) Pr X 1 x1, X 2 x2 ,..., X k xk
On the pair X 1 , X 2 of discrete random variable, sometimes we interested in its
marginal probabiliy of X 1 or X 2 . The following is the definition of marginal
pdf of X 1 and X 2 .
Definition 5.3
If the pair X 1 , X 2 of discrete random variable has the joint pdf f ( x1 , x2 ,..., xk ) ,
then the marginal pdfs of X 1 and X 2 are
f1 ( x1 ) f ( x1 , x2 )
x2
And
f 2 ( x2 ) f ( x1 , x2 )
x1
Example :
let
random
f ( x1 , x2 )
variables
X 1 and
X2
having
joint
density
function
x1 x2
, x1 1, 2, 3 and x2 1, 2
21
21
21
x2 1
2
f1 ( x1 ) f ( x1 , x2 )
x2
21
21
x1 1
3
f 2 ( x2 ) f ( x1 , x2 )
x1
xk
x1
...
x2 x1
f (t , t )dt t
1
1 2
0 0
x2 x1
4t t dt t
12
1 2
x12 x22
0 0
the pair
X 1 , X 2 of
f ( x1 , x2 )dx2
And
f 2 ( x2 )
f ( x1 , x2 )dx1
Example :
By using the previous example, marginal density function of X 1 is
1
f1 ( x1 ) 4 x1 x2 dx2 4 x1 x2 dx2 2 x1
5.3 Independent Random Variabel
As in chapteer 2 we have discussed independent events. The concept on those
independence caan be extended in random variable. The following theorem
variable X and Y
1
for
4
1
f X ( x) f ( x, y ) , for x 1,1 and
2
y
1
fY ( x) f ( x, y ) , for x 1,1
2
x
f X ( x ) fY ( x ) f ( x , y )
1
, then X and Y independent.
4
Definition 5.6
Random variable X and Y are called identically distributed if X and Y have equal
distribution distribution , that is FX ( x) FY ( y )
Definiton 5.7
Random variables X 1 , X 2 ,..., X k said to be independent and identically distributed
(iid) if and only if
This properties of random variables have important role in sample random that
will discussed in the next discussion
f ( x2 | x1 )
4 x1 x2
0
f ( x1 , x2 )
for 0 x1 1, 0 x2 1
elsewhere
Find the marginal densities of X 1 and X 2 and the contional density of X 1 given
X 2 x2
Solution
f1 ( x1 )
f ( x1 , x2 ) dx2
4 x1 x2 dx2
0
2 x1 x22 |10 2 x1
for 0 x1 1, and f1 ( x1 ) 0 elsewhere ;also
f 2 ( x2 )
f ( x1 , x2 )dx1
4 x1 x2 dx1
0
2 x2 x12 |10 2 x2
for 0 x2 1, and f 2 ( x2 ) 0 elsewhere, then substituting into the formula for
conditional density, we get
f ( x1 | x2 )
f ( x1 , x2 ) 4 x1 x2
2 x1
f 2 ( x2 )
2 x2
X 1 , X 2 ,..., X k
Fk ( X )
k
1
x y1
yi x yi 1
yk x
We have learnt the use of random variable and its probability distribution as a way
for expressing mathematical models of nondeterministic event. Next, we will
discuss the extended of properties and some descriptive measurement involving
more than one random variable.
5.6 Properties of Expected Value
Theorem 5.2
If X ( X 1 , X 2 ,..., X k ) has a joint pdf
If X is discrete
u X ,..., X f x , x ,..., x
1
x2
E X u X 1 ,..., X k
u X ,..., X f x ,..., x dx K dx
1
If X is continuous.
Theorem 5.3
If X 1 and X 2 are random variables with joint pdf f ( x1 , x2 ) , then
E X1 X 2 E X1 ) E ( X 2
Theorem 5.4
If X and Y independent random variables , g(x) and h(y) are function then
E g ( x ) h( y ) E g ( x )) E ( g ( y )
Definition 5.11
The Covariance of a pair random variable X and Y is defined by
Cov ( X , Y ) E X X Y Y
Theorem 5.6
If X and Y are random variables, then
Cov ( X , Y ) E ( XY ) E ( X ) E (Y )
Proof:
Var ( X Y ) E X Y 1 2
E X 1 Y 2
2
2
E X x E Y y 2 E X x Y y
Definition 5.11
2
2
If X and Y are random variables with variances X and Y and covariance
XY
XY
And
E Y | x
yf ( y | x)dy
if 0 y
x
2
f1 ( x) dy
0
f ( y | x)
E (Y | x )
x
2
f ( x, y ) 2
f1 ( x)
x
x
1
2
for 0 x 0
for 0 y
x
2
yf ( y | x )dy
x /2
E (Y | x)
2
x
y dy
x
4
for 0 x 2
Theorem 5.13
If X and Y are random variable have joint distribution then
E (Y | X ) E (Y )
Proof:
E (Y | X )
E (Y | x) f ( x)dx
yf ( y | x) f ( x)dydx
y f ( x, y )dxdy
yf ( y)dy E (Y )
Definiton 5.13
The conditional variance of Y given X x is given by
Var (Y | x) E Y E (Y | x ) | x
Example:
If X and Y are random variable with joint density function
f ( x, y )
4
if x 1, 2 and y 2,3 then
5 xy
3
4
20
f ( x, y ) 12
E
(
Y
|
x
)
and
30 x
f ( x)
5
y 2 5 xy
y 2
f1 ( x )
Var Y | x E Y 2 | x E Y | x
3
y2
y 2
f ( x, y ) 12
6
f ( x) 5
25
Theorem 5.14
If X and Y random variables with joint distribution then
Var (Y ) E X Var Y | X VarX E Y | X
proof :
2
E X Var Y | X E X E Y 2 | X E Y | X
EY EY E EY | X EY
E Y 2 EX E Y | X
2
Var (Y ) VarX E Y | X
Theorem 5.15
If X and Y are random variables with joint distribution and h(x,y) are function
then
E h( X , Y ) E X E h ( X , Y ) | X
Theorem 5.16
If X and Y are random variables with joint distribution and g(x) are function
then
E g ( X )Y | x g ( x ) E Y | x
5.8
Exercise 5
1. Let X 1 , X 2 , X 3 and X 4 be independent random variables, each having the
same distribution with mean 5 and standard deviation 3, and let
Y X1 2 X 2 X 3 X 4
a. Find E (Y )
b. Var (Y )
2. Suppose X and Y are continuous random variables with joint pdf
f ( x, y ) 24 xy if 0 x, 0 y and x y 1 , and zero otherwise
a. Find E ( XY )
b. Cov( X , Y )
c. Find correlation coefficient of X and Y
d. Find Cov(3 X ,5Y )
e. Find Cov ( X 1, Y 2)
3. Let f ( x, y ) 6 x , 0 x y 1 , and zero otherwise, find:
a. f1 ( x )
b. f 2 ( y )
c. Cov( X , Y )
d.
e. f ( y | x)
f. E (Y | x )
4. Let X and Y have joint pdf f ( x, y ) e y if 0 x y and zero
otherwise. Find E ( X | y )
CHAPTER VI
FUNCTION OF RANDOM VARIABLE
FY ( y ) Pr Y y
Pr e x y
Pr X ln y
FX ln y 1 y 2
1 y
d
FY ( y ) 2 y 3
dy
1 y
fY ( y ) f X w( y )
yB
Where B y | fY ( y ) 0
Example:
Let X ~ GEO ( p ) so that
f X ( x) pq x 1
x 1, 2,3,...
y 0,1, 2,...
b. Continuous Case
Suppose that X is a continuous random variable with pdf f X ( x ) and assume
that Y u ( x ) defines a one-to-one transformation from A x | f X ( x) 0 on
to B y | fY ( y ) 0 with inverse transformation x w( y ) . If the derivative
fY ( y ) f X ( w( y ))
d
w( y )
dy
yB
Example;
2 x
Suppose that FX ( x ) 1 e , 0 x , if Y e x , find density function of Y.
Solution:
f X ( x) 2e 2 x , the term of its transformation inverse x w( y ) ln y , then
d
1
ln y
dy
y
fY ( y ) f X ( w( y ))
2e 2ln y
d
w( y )
dy
2 1
1
2eln y
2 y 3 ,1 y
y
y
u ( x) , the
Suppose f X ( x )
4 1
, x 2, 1, 0,1, 2 , and let Y | X | . Partition A into
31 2
xj
On A1
u( x j ) y
0
1
2
0
1
2
xj
On A2
u( x j ) y
-1
-2
fY (0) f X ( x j ) f X (0)
j
1
2
4
31
4 1
31 2
4 1
31
2
4 1
4 1
fY (2) f X ( x j ) f X (2) f X (2)
31 2
31
2
j
b. Continuous Case
In continuous random variable case, density function can be found as in
discrete case, thus , its density function
dx
d
fY ( y ) f X ( w j ( y ))
w j ( y ) or fY ( y ) f X ( x j ) j
dy
dy
j
j
Example:
Suppose that X ~ UNIF (1,1) and Y X 2 . If we partition A (1,1) into
A1 (1,0) and A2 (0,1) , then y x 2 has unique solution x1 w1 ( y ) y
and x2 w2 ( y ) y over this interval. We can neglect the point x 0 in this
partition, because X is continuous. The pdf of Y is thus
1
1
1
fY ( y ) f X ( y )
fX ( y)
y B (0,1)
2 y
2 y 2 y
6.2.3 Joint Transformation
a. Discrete Case
If X is a vector of discrete random variables with joint pdf f X ( x) and
Y u ( X ) defines a one-to-one transformation, then the joint pdf of Y is
fY ( y1 , y2 ,..., yk ) f X ( x1 , x2 ,..., xk )
x1 x2
, x1 1, 2,3 and
36
y1
and
y2
y1
y
, y 2 1 , y1 1, 2,3, 4, 6,9 . Marginal
36
y2
x2 y 2 , therefore, fY ( y1, y 2 ) f X
density function of Y1 can be easy found by using the following table at first.
Tabel 2: the values of y1 , y2 of values x1 and x2
x1
1
2
3
y1 x1x2
1
2
3
2
4
6
3
6
9
x2
1
2
3
1
2
3
1
2
3
fY1 ( y1 )
y2
y2 x 2
1
2
3
1
2
3
1
2
3
y1
36
Thus,
for y1 1then y2 1 fY1 (1)
1
36
2
2
4
36 36 36
4
36
3 3
6
36 36 36
6
6 12
36 36 36
9
36
Or
y1
36 ,
fY1 ( y1 )
2 y1 ,
36
untuk y1 1, 4,9
untuk y1 2,3, 6
b. Continuous Case
let X is a vector of continuous random variables with joint pdf f X ( x) 0 over
A and
x1
y2
...
x1
yk
x2
J y1
M
xk
y2
x2
y2
...
x2
yk
M O
xk
...
y2
M
xk
yk
Example:
Consider random variables X 1 and X 2 , then defined new random variables
Y1 X 1 and Y2 X 1 X 2 .
x1
y1
x2
y1
x1
y2
1
y2
x2
y12
y2
y2
, its jacobian is
y1
0
1
1
y1
y1
1
.joint density function of X 1 and X 2 are f x1, x2 4 x1x2 ,
y1
0 x1 1 and 0 x2 1 , then
fY ( y1, y2 ) f X ( x1, x2 ) J
y 1
f X y1, 2
y1 y1
y 1
y
4 y1 2
4 2
y1 y1
y1
0 y1 1 and 0 y2 1
6.3 MGF Method
Theorem 6.1
MGF of Y X i is
i 1
M Y (t ) M X1 (t )M X 2 (t )...M X k (t )
Example:
Let X 1 , X 2 ,..., X k be independent binomial random variables with repective
k
M Y (t ) M X1 (t ) M X 2 (t )...M X k (t )
pe
pe q
pet q
t
n1
n2
... pet q
nk
n1 n2 ... nk
We recognize that this is the binomial MGF with parameters n1 n2 ... nk and
p , and thus Y ~ BIN (n1 n2 ... nk , p )
pdf, f ( x) , where
Yk is given by
g k ( yk )
n!
k 1
n k
F ( yk ) 1 F ( yk ) f ( yk )
(k 1)!(n k )!
n!
hi , j ( yi , y j ) with
(i 1)!( j i 1)!(n j )!
hi , j ( yi , y j ) F ( yi )
i 1
f ( yi ) F ( y j ) F ( yi )
j i 1
1 F ( y j )
n j
f (yj)
a yi y j b.
Special case:
The density function of order statistic minimum is
g1 ( y1 ) n 1 F ( y1 )
n 1
f ( y1 ) where a y1 b,
n 1
f ( yn ) where a yn b,
Y1 min X 1, X 2 ,..., X n
Yn max X 1, X 2 ,..., X n
Distribution function of order statistic minimum can be found by
G1 ( y1 ) Pr Y1 y1
1 Pr Y1 y1
1 Pr every X i xi
1 1 F ( y1 )
1 Pr Y1 y1
1 Pr every X i xn
F ( yn )
Example:
Suppose X 1 , X 2 , X 3 are random variable with size 3of population with density
function f ( x ) 2 x 0<x<1.
Joint density function of order statistic Y1 , Y2 , Y3 is
g ( y1, y2 , y3 ) 3! f ( y1 ) f ( y2 ) f ( y3 )
48 y1 y2 y3
0 y1 y2 y3 1
x
F ( x) 2tdt x 2
0
g1 ( y1 ) n 1 F ( y1 )
g n ( yn ) n F ( yn )
n 1
n 1
f ( y1 ) 6 y1 1 y12
Excercise 6
for 0 y1 1
f ( yn ) 6 yn5 for 0 yn 1
Gn ( yn ) F ( yn ) yn6 for 0 yn 1
n
X1 X 2
X1
X2
5. Consider a sample random of size n from an exponential dsitribution
b. Find the pdf of W
CHAPTER VII
STATISTIC AND SAMPLING DISTRIBUTION
Definition 7.2
Xi
1 n
2
S
Example:
Let
sample mean and the sample variance are statistic cause both of them do not
depend on value of and 2 . If taken T X then T is not statistic because T
contain unknown . However if the value of known (example =2) then T is
statistic.
The following theorem provides important properties of the sample mean and
sample variance
Theorem 7.1
If
Var X 2 , then
1. E ( X )
2
n
2
3. Var ( S ) 2
2. Var ( X )
Proof:
n
X i
1
1
1. E ( X ) E E Xi n
n i 1
n
i 1 n
n
n
X
1
2. Var ( X ) Var i Var
Xi
i 1 n
n i 1
n
2 Var
n
1
2
2
X i 2 n
n
n
i 1
3. Var ( X ) E X
E X
E ( X ) and
2
2 , thus
n
2
1
1 n
Var ( S 2 ) E
Xi X
E
n 1
n 1 i 1
2
1 n 2
E X i E n X
n 1 i 1
then E X
X
i 1
2
i
2
nX
1
2
2
2
n
(
n
(
2 )
n 1
n
1
n 2 n 2 2 n 2 )
n 1
1
2 (n 1) 2
n 1
2
X
,
X
,...,
X
Let 1 2
n are sample random obtained on distribution N ( , ) , then we
Consider Y
E (etY ) E e n
t
X1 X 2 ... X n
Ee
t
n
X1
e n 2 ...e n
Xn
E e n 1 e n 2 ...e n
E e E e ...E e
X
E e M (t ) for Z
.
n
t
n
Xn
t
n
Noticed that
t
n
X1
t
n
X2
Xi
Xn
n
t t
M Z (t ) M X e n 2 n2 , and M Y (t ) M X i
n
i 1
22
t
, since
n
n
X 1 , X 2 ,..., X n are
t 2t 2
2
t
t
M
exp
exp
identic thus, X i
Xi
2
n
n
n
i 1
n 2n
t2
2 t 2
2
2
M Y (t ) exp t
N
,
X
~
N
,
therefore
that is MGF of
n
n
n 2
Theorem 7.2
If Z ~ N (0,1) then Z 2 ~ 2 (1)
Proof:
M Z 2 (t ) E e tz
1 tz 2 z22
e
2
1 z2
e
2
1 2t
1 2t
dz
1 z2
e
2
1
1 2t
1 2 t
2
1
e
2
1 2 t
2
1 2 t z
2
1 2t z
1 2t z
12
i 1
V Yi ~ 2
v
i
i 1
Corollary 7.1
Let X 1 , X 2 ,..., X n denotes a sample random from N ( , 2 ) then
n
X i ~ 2 ( n)
a.
2
i 1
b.
n X
~ 2 (1)
Proof
a. Let Yi
Xi
then
M Yi (t ) e
t M t e t e
X
t 2t 2
2
t2
e2
2
2
It means Yi ~ N (0,1) consequence Yi ~ (1) , thus,
Y
i 1
~ 2 ( n)
n X
b. Suppose Y
2
n
n . We know that X ~ N ,
X
n
2 t2
and M X (t ) e t n 2 consequence
M Y (t ) e
n
t
n
M X
t
n
t
n
2 2 n
t t
n 2 2
t2
2
e
e
e
Y
~
N
(0,1)
It means
therefore Y 2 ~ 2 (1)
( n 1) S 2
, first notice the following terms
2
X
i 1
i 1
i 1
X 2 X i 2
2
2
i
i 1
n
X i2 2n X n 2
i 1
2n X n
X i2 n X n X 2n X n 2
X i2 n X n X
i 1
i 1
n
i 1
i 1
X i2 X X i n X n X 2n X n 2
n
Xi X
i 1
n X
Thus,
n
i 1
Xi
i 1
i X
1 n
But, S
Xi X
n 1 i 1
2
i 1
Xi
2
n X
(n 1) S 2 n X
2
2
Or V1 V2 V3
2
2
As corrolary 7.1 we have V1 ~ (n) and V3 ~ (1) , since V2 and V3 independent
then
Ee e
Ee Ee
M V1 (t ) E e t (V2 V3 )
tV2 tV3
tV3
tV2
M V2 (t ) M V3 (t )
Therefore, M V2 (t )
M V1 (t )
M V3 (t )
1 2t
M V (t )
1 2t
n2
1
2
1 2t
2
Therefore, V2 ~ (n 1) so
n2 12
1 2t
( n21)
( n 1) S 2
~ 2 (n 1)
2
Z
V is called t-Student distribution with degree of freedom v and notated by
v
v 1
2
f (t ; v )
v
2
The
table
of
1
v
t
1
v
( v21)
distribution
Pr(T t ) 1 Pr(T t ) .
, t ;0 v
gives
value
of
Pr(T t ) ,
and
The value of t (v) states the value of percentile that staisfies Pr(T t (v)) .
Example:
If T ~ t (17) , determine Pr(T 1.74)
Solution:
t table gives value of P(T t ) then
Pr(T 1.74) 1 Pr(T 1.74)
1 0.95 0.05
V1
v1
V2
v2
is
1
~ F (v2 , v1 )
X
Pr X f (v1 , v2 )
1
1
f (v1 , v2 )
X
Pr
1
1 Pr Y
f (v1 , v2 )
1
Then 1 Pr Y
f (v1 , v2 )
1
Thus, Pr Y
Pr Y f1 (v2 , v1 )
f (v1 , v2 )
So,
1
f1 (v2 , v1 )
f (v1 , v2 )
Example:
1.
f 0.1 (15, 7)
1
1
0.463
f 0.1 (7,15) 2.16
1
0.25 Pr X 4 0.975
X
2. If X ~ F (20,8) then Pr
Excercise 7.
b. Pr Z i Z 25
i 1
T
~
t
(
v
)
3. Let
find the distribution of T 2
2
4. If X i ~ N ( , ), i 1, 2,..., n and Z i ~ N (0,1), i 1, 2,..., k , determine
n X
distribution of
SZ
distribution of k Z
6. If T ~ t (3) calculate P (0.87 T 2.65)
7. If X ~ F (5,10) determine the value of a and b , then
a. Pr( X a ) 0.05 and Pr( X b) 0.95
b. Pr(a X b) 0.9
REFERENCE
th