2.
Random variables
Introduction
Distribution of a random variable
Distribution function properties
Discrete random variables
Point mass
Discrete uniform
Bernoulli
Binomial
Geometric
Poisson
2. Random variables
Continuous random variables
Uniform
Exponential
Normal
Transformations of random variables
Bivariate random variables
Independent random variables
Conditional distributions
Expectation of a random variable
kth moment
2. Random variables
Variance
Covariance
Correlation
Expectation of transformed variables
Sample mean and sample variance
Conditional expectation
Introduction
Random variables assign a real number to each
outcome:
X :
X ( )
Random variables can be:
Discrete: if it takes at most countably many
values (integers).
Continuous: if it can take any real number.
RANDOM VARIABLES
Distribution of a random variable
Distribution function
F ( x) FX ( x) P ( X x)
RANDOM VARIABLES
Distribution function properties
(i)
F ( x) 0 when
(ii)
F ( x) 1 when
(iii) F (x)
is nondecreasing.
x1 x2 F ( x1 ) F ( x2 )
(iv) F (x)
is right-continuous.
F ( x) F ( x0 ) when
x x0
x x0
RANDOM VARIABLES
Distribution of a random variable
For a random variable, we define
Probability function
Density function,
depending on wether is either discrete or continuous
RANDOM VARIABLES
Distribution of a random variable
Probability function
p ( x) p X ( x) P( X x)
verifies
(i ) p ( x) 0
(ii ) p ( x) 1
x
RANDOM VARIABLES
Distribution of a random variable
Probability density function
f (x)
verifies
(i )
(ii )
f ( x) 0
f ( x)dx 1
We have
x
F ( x) f (t )dt and f ( x) F ' ( x).
RANDOM VARIABLES
Distribution of a random variable
F completely determines the distribution
of a random variable.
p( x)
a x b
P (a X b) F (b) F (a )
f (t )dt
RANDOM VARIABLES
10
Discrete random variables
Point mass
X a
P( X a) 1
0 if
F ( x)
1 if
1--
xa
xa
RANDOM VARIABLES
11
Discrete random variables
Discrete uniform
X U (1,2,..., k )
1
P( X i)
i 1,2,..., k
k
k-1
RANDOM VARIABLES
12
Discrete random variables
Bernoulli
X B (1, p )
P ( X 1) p
P ( X 0) 1 p
p
p
1-p
1-p
0
RANDOM VARIABLES
13
Discrete random variables
Binomial
Successes in n independent Bernoulli trials with
success probability p
X B ( n, p )
n x
P ( X x) p (1 p ) n x
x
n
n!
with
x!(n x)!
x
x 0,1,2,..., n
RANDOM VARIABLES
14
Discrete random variables
Geometric
Time of first success in a sequence of independent
Bernoulli trials with success probability p
X G ( p)
x 1
P ( X x) (1 p ) p
x 1,2,3,...
RANDOM VARIABLES
15
Discrete random variables
Poisson
X expresses the number of rare events
X P ( ), 0
e
P( X x)
x!
x 0, 1, 2,...
RANDOM VARIABLES
16
Continuous random variables
Uniform
X U [ a, b]
for a x b
f ( x) b a
0 otherwise
0 for x a
x a
F ( x)
for a x b
ba
1 for x b
F(x)
f(x)
b
RANDOM VARIABLES
17
Continuous random variables
Exponential
1 x
for x 0
e
f ( x)
0 for x 0
0 for x 0
x
F ( x)
e
for x 0
X exp( )
1/
1
F(x)
f(x)
RANDOM VARIABLES
18
Continuous random variables
Normal
X N ( , 2)
( x )2
1
f ( x)
exp
2
2
2
2 0
f(x)
F(x)
RANDOM VARIABLES
19
Continuous random variables
Properties of normal distribution
(ii)
X
N (0,1) standard normal
Z N (0,1) Z N ( , 2 )
(iii)
X i N ( i , i2 )
(i)
independent i=1,2,...,n
n
X i N ( i , )
i
2
i
RANDOM VARIABLES
20
Transformations of random variables
X random variable with
FX
Y = r(x); distribution of Y ?
r() is one-to-one; r -1().
1
FY ( y ) P(Y y ) P(r ( X ) y ) P( X r ( y )) FX (r ( y ))
pY ( y ) P(Y y ) P(r ( X ) y ) P( X r 1 ( y )) p X (r 1 ( y ))
fY ( y )
d
dy
FX (r ( y )) f X (r ( y ))
d r 1 ( y )
dy
RANDOM VARIABLES
21
Bivariate random variables
(X,Y) random variables;
If (X,Y) is a discrete random variable
p( x, y ) probability joint function
verifies : p ( x, y ) 0
p( x, y) 1
x, y
If (X,Y) is continuous random variable
f ( x, y ) probability density joint function
verifies : f ( x, y ) 0
f ( x, y)dxdy 1
RANDOM VARIABLES
22
Bivariate random variables
The marginal probability functions for X and Y are:
p X ( x ) p ( x, y )
y
pY ( y ) p ( x, y )
x
For continuous random variables, the marginal
densities for X and Y are:
f X ( x) f ( x, y )dy
fY ( y ) f ( x, y )dx
RANDOM VARIABLES
23
Independent random variables
Two random variables X and Y are independent if
and only if:
p ( x, y ) p X ( x) pY ( y )
f ( x, y ) f X ( x) fY ( y ),
for all values x and y.
RANDOM VARIABLES
24
Conditional distributions
Discrete variables
p ( x, y )
p( x | y ) P( X x | Y y )
p( y )
Continuous variables
f ( x, y )
f ( x | y)
f ( y)
If X and Y are independent:
p ( x | y ) p ( x)
f ( x | y ) f ( x)
RANDOM VARIABLES
25
Expectation of a random variable
EX X xp( x)
x
EX X xf ( x)dx
Properties:
(i) E
X i i E X i
i 1,..., n
(ii) If X i , i 1,..., n are independent then:
E X i EX i
i
RANDOM VARIABLES
26
Moment of order k
EX x p ( x)
k
EX x f ( x)dx
k
RANDOM VARIABLES
27
Variance
Given X with
EX
VX E ( X )
2
X
X VX ( E ( X ) )
2 1/ 2
standard deviation
RANDOM VARIABLES
28
Variance
Properties:
(i)
V (aX b) a V ( X )
2
(ii) If X i are independent then
V ( ai X i ) ai V ( X i )
2
(iii)
VX EX (EX )
(iv)
VX 0
VX 0 P ( X a) 1
RANDOM VARIABLES
29
Covariance
X and Y random variables;
Cov ( X , Y ) E ( X EX )(Y EY )
Properties
(i) If X, Y are independent then
(ii)
cov( X , Y ) 0
Cov ( X , Y ) EXY EXEY
(iii) V(X + Y) = V(X) + V(Y) + 2cov(X,Y)
V(X - Y) = V(X) + V(Y) - 2cov(X,Y)
RANDOM VARIABLES
30
Correlation
X and Y random variables;
Cov ( X , Y )
( X ,Y )
VX VY
RANDOM VARIABLES
31
Correlation
Properties
(i) 1 ( X , Y ) 1
(ii) If X and Y are independent then ( X , Y ) 0
(iii)
( X , Y ) 1 a 0 : Y aX b
( X , Y ) 1 a 0 : Y aX b
RANDOM VARIABLES
32
Expectation of transformed variables
Y r ( X );
Er ( X ) r ( x) p X ( x)
x
Er ( X ) r ( x) f X ( x)dx
RANDOM VARIABLES
33
Sample mean and sample variance
Sample mean
1
EX X X i
n i
Sample variance
1
2
V (X ) S
(Xi X )
n 1 i
2
RANDOM VARIABLES
34
Sample mean and sample variance
Properties
X random variable; EX , VX ;
X 1 ,..., X n i. i. d. sample,
2
Then:
EX
2
(ii) VX
n
(iii) ES 2 2
(i)
RANDOM VARIABLES
35
Conditional expectation
X and Y are random variables; X | Y y.
Then:
E ( X | Y y ) x p( x | Y y )
x
E ( X | Y y ) x f ( x | y )dx
Properties:
EE ( X | Y ) EX
RANDOM VARIABLES
36
