You are on page 1of 7

Section 3.

2: Probability Distributions for


Discrete Random Variables

Section 3.2 introduces the following concepts for


a discrete random variable X
Probability mass function (PMF)
p(a) = P (X = a)
for a possible value a.
Parameters: if the PMF contains unknown
members, then we call those unknown members as parameters.
Cumulative Distribution Function (CDF),
F (a) = P (X a)
for any real number a.
A CDF is nondecreasing, right-continuous,
lim F (a) = 0

and
lim F (a) = 1.

If the PMF is given, the CDF can be computed by


F (a) =

p(t).

ta

If the CDF is given, the PMF can be computed by


p(a) = F (a) lim F (t) = F (a) F (a).
ta

If X are integer-valued, then we have


p(a) = F (a) F (a 1).

First example of Section 3.2: example 3.9.


Assume 80% customer purchase a laptop and
20% purchase a desktop
Suppose X = 1 if a customer purchases a laptop and 0 if the customer purchases a desktop.
Then, the PMF is given as
p(1) = P (X = 1) = 0.8
p(0) = P (X = 0) = 0.2
and
P (X = other value) = 0.
It can also expressed as a table
a
p(a)

0
0.2

1
0.8

Second example of Section 3.2: example 3.10.


Consider the following PMF table

a
p(a)

1
0.4

2
0.3

3
0.2

4
0.1

The CDF is

0.4

F (a) =

0.7

0.9

if
if
if
if
if

a<1
1a<2
2a<3
3a<4
a4

Third example of Section 3.2: example 3.12.


Suppose P (S) = p and P (F ) = 1 p.
Independently repeat the experiment until S
appears.
Let X be the number of runs of experiment.
Then, the PMF is given by

p(a) =

(1 p)a1p,
0

if a = 1, 2,
otherwise.

Fourth example of Section 3.2: examples 3.15.


Suppose the CDF is given as

F (a) =

0.58

0.72

0.76

0.81

0.88

0.94

...

a<0
0a<1
1a<2
2a<3
3a<4
4a<5
5a<6

Then, we have
P (2 X 5) = F (5) F (1) = 0.94 0.72 = 0.22
P (2 < X 5) = F (5) F (2) = 0.94 0.76 = 0.18
P (2 X < 5) = F (4) F (1) = 0.88 0.72 = 0.16
and
P (2 < X < 5) = F (4)F (2) = 0.880.76 = 0.12.

You might also like