Lec 05

2.4 Cumulative Distribution Function (CDF)

The cumulative distribution function of random variable X is

F X ( x)  P[ X  x]

For any real number x, the CDF is the probability that the random variable
X is no larger than x.

For all b  a

F X (b)  F X (a)  P[a  X  b]

X : Subscript corresponds to the name of the random variable

Properties of Cumulative Distribution Function (CDF)

For any discrete random variable X with range S X  {x1 , x2 ,...} satisfying
x1  x2  ... ,

a) F ()  0 and F ()  1 .

b) For all x '  x, FX ( x ')  FX ( x) .

c) For xi  S X and  , and arbitrarily small positive number

FX ( xi )  FX ( xi   )  PX ( xi )

d) FX ( x)  FX ( xi ) for all x such that xi  x  xi  1

1
Ex 2.23] Random variable R and PMF given as

1
4 r  0

3
PR (r )   r2
4
0 otherwise



Plot PMF and corresponding CDF

0 .8

0 .6
PR(r)

0 .4

0 .2

0
-1 -0 . 5 0 0 .5 1 1 .5 2 2 .5 3
r

0 r  0
1

FR (r )   0r 2
 4
1 r  2

2
Ex 2.24] Let’s consider the geometric  p   random variable, Y with PMF
1
 4
given below

 1   1  y 1  1  3  y 1
  1       y  1, 2, 3,...
 4   4   4  4 

PY [ y ]  
0 otherwise



What is cumulative distribution function (CDF)?

i) when y  
ii) when y  n


j 1
n
 1  3 
FY  n       
j 1  4   4 

y then

Let Z  FY  n  ,
j 1 0 1 2 3
n
 1  3  1  3   1  3   1  3   1  3 
Z                         ...
j 1  4  4  4  4   4  4   4  4   4  4 

For this kind problem, the geometric series can be used to get solutions.

 0
 y y 1
 3
FY  y   1    y 1
  4 
y
 1


3
Q 2.4] Use CDF to find the following probabilities and draw the PMF plot

CDF of F[y]

0.8

0.6
F(y)

0.4

0.2

-0.2
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
y

(1) P[Y < 1] = FY (1−) = 0

(2) P[Y ≤ 1] = FY (1) = 0.6

(3) P[Y > 2] = 1 − P[Y ≤ 2] = 1 − FY (2) = 1 − 0.8 = 0.2

(4) P[Y ≥ 2] = 1 − P[Y < 2] = 1 − FY (2−) = 1 − 0.6 = 0.4

(5)P[Y = 1] = P[Y ≤ 1] − P[Y < 1] = FY (1+) − FY (1−) = 0.6

(6) P[Y = 3] = P[Y ≤ 3] − P[Y < 3] = FY (3+) − FY (3−) = 0.8 − 0.8 = 0

4
2.5 Averages

Mean, Median, and Mode

Ex] 9 5 10 8 4 7 5 5 8 7

Median: 7

Mode: 5

Mean: 68/10 = 6.8

Mean is Expected value.

The expected value of X is defined as

E[ X ]   xP
xS X
X ( x)

5
  1
The geometric (p) random variable X has expected vale  E  X   
p  

Proof: Let  q  1  p  . The PMF of X becomes

 pq x 1 x  1, 2, 3,...
PX ( x)  
0 otherwise

 
p 
E[ X ]   xPX ( x)   xpq x 1   xq x
x 1 x 1 q x 1

Math fact B.6 in page 508

n
q
If q  1 ,  iq
i 1
i

(1  q ) 2

So,
 
p 
E[ X ]   xPX ( x)   xpq x 1
  xq x
x 1 x 1 q x 1
p q p q
=  = 
q (1  q) 2
q (1  q) 2
p p
=  2
(1  q) 2
p
1
=
p

For discrete uniform (k, l) random variable X, the mean value is

k l
EX  
2