Vous êtes sur la page 1sur 28

Chapter 2: Random Variables

2.1 Introduction
X => a random variable; why random?
Outcome of experiment of chance
Also known as a function X defined on a prob. space

Outcome quantitative or qualitative


X => age of a person drawn randomly from one pop.
X => eye color of a person drawn randomly from one pop.
Other examples?
Details on discrete and continuous r.v.
1 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
Chapter 2: Random Variables
2.2 Discrete random variables
A discrete sample space, => a discrete r.v.
If X is a discrete r.v., the distribution of X is given
by the probability function (p.f.)
Notation used:
The p.f. has the following properties:
2 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( ) 0 >
i
x f
( ) 1
1
=

i
x f
Prob X takes a
specific value
Chapter 2: Random Variables
A p.f. can takes the form --- formula or a table
listing all the possible values of x and its
probabilities, f(x).
Suppose we flip a coin two times. The sample space
of this experiment is {HH, HT, TH, TT}. Suppose X is a
random variable, denoting the number of heads.
Based on the sample space, we know that X can
take values of 0, 1 or 2.
A p.f. in terms of formula is:
3 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
x x
x
x f

|
|
.
|

\
|
=
2
5 . 0 5 . 0
2
) (
Chapter 2: Random Variables
Can easily verify that all f(x) for x = 0,1,2 is greater
than or equal to 0 and that the sum of f(x
i
) is 1
A p.f. can also be described in a table
A p.f. relates each value of a discrete r.v. with its
probability of occurrence.
4 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
Number of Heads (X) Probability (f(X))
0 0.25
1 0.50
2 0.25
Obtained by looking
at the sample space
Chapter 2: Random Variables
A vector of two discrete r.v (X and Y) is defined
by the joint p.f.
Notation used:
Usually a joint p.f. is given in a two-way table.
Suppose X could take these values {S, H, P, C} and Y
could take the values {0, 1}
5 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( ) ( ) y Y x X P y x f = = = , ,
S H P C
0 3/52 3/52 3/52 3/52
1 10/52 10/52 10/52 10/52
P(X=S and Y=1) = 10/52
Sum => 1
Chapter 2: Random Variables
Marginal distribution => distribution of the
subset of random variables
Example of marginal dist: Suppose there is a
random vector (X, Y) with the joint p.f. f(x,y).
Thus, we will have univariate marginal distributions
6 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( )

=
y
X
y x f x f ) , (
( )

=
x
Y
y x f y f ) , (
Sum all the values for each
given Y
Sum all the values for each
given X
Chapter 2: Random Variables
7 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
S H P C
0 3/52 3/52 3/52 3/52
1 10/52 10/52 10/52 10/52
Y
X
Based on the above joint p.f. find the marginal distribution of X
and marginal distribution of Y
Chapter 2: Random Variables
If there is a random vector (X,Y,Z) with the joint
p.f. f(x,y,z) => what are the marginal
distributions that one can consider?
Need to think of all possible subsets
There will be univariate and bivariate
8 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( )

=
z y
X
z y x f x f
,
) , , ( ( )

=
z
Y X
z y x f y x f ) , , ( ,
,
Other possible examples?
Chapter 2: Random Variables
2.3 Distribution functions
Mathematical notation is:
Accumulated probability from - up to x
Also known as cumulative distribution function,
c.d.f.
The properties of c.d.f are given as follows:
For all x =>
F is nondecreasing
F is right-continuous (not necessarily strictly)
9 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( ) ( ) x X P x F s =
( ) 1 0 s s x F
Chapter 2: Random Variables
10 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
Example of c.d.f.
Discrete
Continuous
Mixed
1
1
1
0
0
0
Chapter 2: Random Variables
The probability that X lies in the interval (a,b],
where a < b is given by:
When X is discrete: The c.d.f. can be obtained
from p.f.
11 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( ) ( ) ( ) a F b F b X a P
X X
= s <
( ) ( )

s
=
x u
u f x F
Chapter 2: Random Variables
Suppose that a random variable X has the
following probability function (p.f.)
Find the following values:
Plot the c.d.f.
12 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
X Probability (f(X))
0 0.4
1 0.3
2 0.2
3 0.1
( ) ) 3 ( ), 2 ( ), 1 ( , 0 s s s s X F X F X F X F
Chapter 2: Random Variables
A median of X is any number m such that:
The median is uniquely defined only if the c.d.f.
is strictly monotonic.
13 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( ) 2 / 1 = s m X P ( ) 2 / 1 = > m X P
&
f (x)
x
Is this monotonic???
Chapter 2: Random Variables
2.4 Continuous random variables
When X is a continuous r.v., the distribution of X
is given by probability density function (p.d.f.)
The properties of p.d.f are:
The c.d.f. is continuous at each x and so prob
assigned to an individual value is 0.
Eg: Suppose the sample space contains all real
numbers, thus P(X= 2.5) = 0 or P(X=-200)=0
14 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
0 ) ( > x f
}


=1 ) ( dx x f
Chapter 2: Random Variables
The relationship between c.d.f (F) and p.d.f. (f)
are given as follows:
Thus, derivative of F (x) will give the p.d.f. (f)
To find the prob of an interval:
An open, closed or half-open intervals have the
same probability
15 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( ) ( ) ( )
}

= s =
x
du u f x X P x F
( ) ( )
}
= < <
b
a
dx x f b X a P
Chapter 2: Random Variables
Graphic presentation of the relationship
between F(x) and f(x)
16 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
Just read the
cumulative
probability F(a)
from the graph
) ( ) ( a X P a F s =
To get F(x), integrate f(x)
Chapter 2: Random Variables
Example: Suppose the c.d.f. of X is given by :-
Find the following probabilities:
P (1 < X < 4) P (X = 3) P (X > 0)
- Find the median of X
17 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( )

>
=

elsewhere 0
0 if 8 . 0 1 x e
x F
x
Chapter 2: Random Variables
Graphic presentation
18 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
Real line
Median: divide
the data sets
into two parts
1
st
quartile:
divide into
another two
parts
3
rd
quartile:
divide into
another two
parts
P = 0.25
P = 0.5
P = 0.75
Chapter 2: Random Variables
Example: Suppose the c.d.f. of X is given by :-
Try plotting the c.d.f. of X
Find the p.d.f. of X (hint: needs to find the derivative)
Based on the obtained p.d.f., find P (0.2 < X < 0.5)
19 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( )

>
< <
s
=
1 1
1 0
0 0
2
x
x x
x
x F
Chapter 2: Random Variables
2.5 Continuous random vectors
Suppose the joint dist of (X, Y) has the following
property:
The joint distribution is called continuous
bivariate dist and can be obtained by:
20 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
Chapter 2: Random Variables
Example of bivariate random variables:
Measure the height and weight of a person drawn
randomly from one population
X => represents the height; Y => represents the
weight
S0 (X, Y) is a random vector having a continuous
bivariate distribution
21 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
Chapter 2: Random Variables
The joint p.d.f. has the following properties:
The marginal distribution can be found by
integrating the joint p.d.f.
22 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( ) ( )
}
=
Y
X
dy y x f x f , ( ) ( )
}
=
X
Y
dx y x f y f ,
Chapter 2: Random Variables
Example: Suppose the p.d.f. is given as f(x, y) =
3(x + y) for x > 0, y > 0 and x + y < 1. Find the
marginal p.d.f. of X.
Based on the formula:
Answer:
23 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( ) ( )
}

+ =
x
X
dy y x x f
1
0
3
(


2
1
3
2
x
Chapter 2: Random Variables
2.6 Conditional distributions
Already discussed this in terms of conditional
probability
Concept is similar and is extended:
The conditional p.f. of X given Y=y is:
Marginal p.f. of X = x is given by this formula:
24 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
Chapter 2: Random Variables
The conditional p.d.f. of X given Y=y is:
Note that, the same notation is used (similar to the
case of p.f.)
Marginal or unconditional p.d.f. of X can be
obtained by:
25 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
Chapter 2: Random Variables
Example: The joint p.d.f. f(x,y) = 2 for 0<y<x<1
and the conditional distribution f(y|x) is
uniform on interval (0, x).
What is the marginal p.d.f. f
X
(x)?
What is the marginal p.d.f. f
Y
(y)?
Obtain the conditional p.d.f. f(x|y).
26 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
Chapter 2: Random Variables
2.7 Independence
Two experiments are independent when the
model does not change with the coming of new
knowledge.
Two r.v. X and Y are independent provided that:
Or alternatively
27 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
Note: f can be either
a p.f. or a p.d.f.
f
Y
(y) cannot be 0
Chapter 2: Random Variables
Example: Suppose that the joint p.d.f. of X and Y
are given as follows:
Verify that both variables X and Y are
independent.
Hint: Demonstrate
28 NAA, Universiti Pendidikan Sultan Idris, Sem1, 2011/12
( ) 0 , 0 for , > > =

y x e y x f
y x
( ) ( ) ( ) y f x f y x f
Y X
. , =

Vous aimerez peut-être aussi