Expectations

Chrysafis Vogiatzis

Department of Industrial and Enterprise Systems Engineering

University of Illinois at Urbana-Champaign

Lecture 9a

©Chrysafis Vogiatzis. Do not distribute without permission of the author

1 / 16 Chrysafis Vogiatzis Expectations

Last time..

Finished with continuous random variables.

We discussed:
Uniform;
Exponential;
Gamma and Erlang;
Normal.
Recall that the normal distribution required two parameters,
referred to as µ and σ 2 ...

2 / 16 Chrysafis Vogiatzis Expectations

Last time..

Finished with continuous random variables.

We discussed:
Uniform;
Exponential;
Gamma and Erlang;
Normal.
Recall that the normal distribution required two parameters,
referred to as µ and σ 2 ...

Today, we will discuss expectations and variances.

2 / 16 Chrysafis Vogiatzis Expectations

Parameters of a distribution

DISCRETE

Name Parameters Values pmf

p(0) = 1 − p
Bernoulli 0<p<1 {0, 1}
p(1) = p
n 
Binomial 0 < p < 1, n ≥ 0 {0, 1, . . . , n} p(x) = px · (1 − p)n−x
x

Geometric 0<p<1 {1, 2, . . .} p(x) = (1 − p)x−1 · p

K
N−K

x n−x
Hypergeometric N, K , n ≥ 0 {1, 2, . . .} p(x) = N

n

λx
Poisson λ>0 {0, 1, . . .} p(x) = e−λ
x!
1
Uniform - [a, b] p(x) =
b−a+1

3 / 16 Chrysafis Vogiatzis Expectations

Parameters of a distribution

CONTINUOUS

Name Parameters Values pdf

1
Uniform - [a, b] f (x) =
b−a

Exponential λ>0 [0, +∞) f (x) = λ · e−λx

λk ·x k −1 ·e−λx
Gamma λ > 0, k > 0 [0, +∞) f (x) = Γ(k )

λk ·x k −1 ·e−λx
Erlang λ > 0, integer k > 0 [0, +∞) f (x) = (k −1)!

1 (x−µ)2
Normal µ, σ 2 (−∞, +∞) f (x) = √ e 2σ 2
2π · σ

4 / 16 Chrysafis Vogiatzis Expectations

Questions
1 What do the parameters reveal about the shape of the
distribution?
2 Knowing the parameters, can we answer questions about what
we should expect will happen?
3 Knowing the parameters, can we answer questions about how far
from the expectation we may find ourselves?

5 / 16 Chrysafis Vogiatzis Expectations

Mean, variance, standard deviation

Specifically, in this class we will provide definitions for:

Mean (or expectation, or expected value) is a measure of the
“center” of the probability distribution, usually denoted by E [X ] or
µ.
Variance is a measure of the variability of the probability
distribution, denoted by Var [X ] or σ 2 .
Standard deviation is another measure of variability, and is
defined as the square root of the variance, denoted by SD [X ] or
σ.
It is called “standard” as it standardizes variability to the same unit
of the original random variable.

6 / 16 Chrysafis Vogiatzis Expectations

Expected values

We separate our discussion between expected values for:

Discrete random variables:
Continuous random variables:

7 / 16 Chrysafis Vogiatzis Expectations

Expected values

We separate our discussion between expected values for:

Discrete random variables:
Definition
Let X be a numerically-valued discrete random variable with sample
space S and probability mass function p(x). Then, the expected
value of X is written as E [X ] and is calculated as:
X
E [X ] = x · p(x).
x∈S

The expected value is commonly referred to as the mean and is also

written as µ.

Continuous random variables:

7 / 16 Chrysafis Vogiatzis Expectations

Expected values

We separate our discussion between expected values for:

Discrete random variables:
P
E [X ] = x · p(x).
x∈S

Continuous random variables:

7 / 16 Chrysafis Vogiatzis Expectations

Expected values
We separate our discussion between expected values for:
Discrete random variables:
P
E [X ] = x · p(x).
x∈S

Continuous random variables:

Definition
Let X be a numerically-valued real random variable defined over
(−∞, +∞) and probability density function f (x). Then, the expected
value of X is written as E [X ] and is calculated as:
+∞
Z
E [X ] = x · f (x)dx.
−∞

The expected value is commonly referred to as the mean and is also

written as µ.

7 / 16 Chrysafis Vogiatzis Expectations

Expected values

We separate our discussion between expected values for:

Discrete random variables:
P
E [X ] = x · p(x).
x∈S

Continuous random variables:

+∞
R
E [X ] = x · f (x)dx.
−∞

7 / 16 Chrysafis Vogiatzis Expectations

Expected values

We separate our discussion between expected values for:

Discrete random variables:
P
E [X ] = x · p(x).
x∈S

Continuous random variables:

+∞
R
E [X ] = x · f (x)dx.
−∞

Recall that:
P
p(x) = 1.
x∈S
+∞
R
f (x)dx = 1.
−∞

7 / 16 Chrysafis Vogiatzis Expectations

How to calculate means
P
Discrete random variables: E [X ] = x · p(x).
x∈S

+∞
R
Continuous random variables: E [X ] = x · f (x)dx.
−∞

Example
A (discrete) random variable is distributed with p(x) = x 2 /c, x = 1, 2, 3, 4.
What is the mean value?
What if the random variable is continuous in [1, 4]?

Answer: Z4 Z4 4
4 4
x3 100 x3 x4 63.75
x ·f (x) = = = .
X X
x · p(x) = = . c 4c c
c c x=1 x=1
1
x=1 x=1

4
P R4
From p(x) = 1, we get that c = 30. From f (x) = 1, we get that c = 21.
x=1 x=1

8 / 16 Chrysafis Vogiatzis Expectations

How to calculate means
P
Discrete random variables: E [X ] = x · p(x).
x∈S

+∞
R
Continuous random variables: E [X ] = x · f (x)dx.
−∞

Example
A (discrete) random variable is distributed with p(x) = x 2 /c, x = 1, 2, 3, 4.
What is the mean value?
What if the random variable is continuous in [1, 4]?

Answer: Z4 Z4 4
4 4
x3 100 x3 x4 63.75
x ·f (x) = = = .
X X
x · p(x) = = . c 4c c
c c x=1 x=1
1
x=1 x=1

4
P R4
From p(x) = 1, we get that c = 30. From f (x) = 1, we get that c = 21.
x=1 x=1

8 / 16 Chrysafis Vogiatzis Expectations

How to calculate means
P
Discrete random variables: E [X ] = x · p(x).
x∈S

+∞
R
Continuous random variables: E [X ] = x · f (x)dx.
−∞

Example
A (discrete) random variable is distributed with p(x) = x 2 /c, x = 1, 2, 3, 4.
What is the mean value?
What if the random variable is continuous in [1, 4]?

Answer: Z4 Z4 4
4 4
x3 100 x3 x4 63.75
x ·f (x) = = = .
X X
x · p(x) = = . c 4c c
c c x=1 x=1
1
x=1 x=1

4
P R4
From p(x) = 1, we get that c = 30. From f (x) = 1, we get that c = 21.
x=1 x=1

8 / 16 Chrysafis Vogiatzis Expectations

How to calculate means
P
Discrete random variables: E [X ] = x · p(x).
x∈S

+∞
R
Continuous random variables: E [X ] = x · f (x)dx.
−∞

Example
A (discrete) random variable is distributed with p(x) = x 2 /c, x = 1, 2, 3, 4.
What is the mean value?
What if the random variable is continuous in [1, 4]?

Answer: Z4 Z4 4
4 4
x3 100 x3 x4 63.75
x ·f (x) = = = .
X X
x · p(x) = = . c 4c c
c c x=1 x=1
1
x=1 x=1

4
P R4
From p(x) = 1, we get that c = 30. From f (x) = 1, we get that c = 21.
x=1 x=1

8 / 16 Chrysafis Vogiatzis Expectations

How to calculate means
P
Discrete random variables: E [X ] = x · p(x).
x∈S

+∞
R
Continuous random variables: E [X ] = x · f (x)dx.
−∞

Example
A (discrete) random variable is distributed with p(x) = x 2 /c, x = 1, 2, 3, 4.
What is the mean value?
What if the random variable is continuous in [1, 4]?

Answer: Z4 Z4 4
4 4
x3 100 x3 x4 63.75
x ·f (x) = = = .
X X
x · p(x) = = . c 4c c
c c x=1 x=1
1
x=1 x=1

4
P R4
From p(x) = 1, we get that c = 30. From f (x) = 1, we get that c = 21.
x=1 x=1

8 / 16 Chrysafis Vogiatzis Expectations

How to calculate means
P
Discrete random variables: E [X ] = x · p(x).
x∈S

+∞
R
Continuous random variables: E [X ] = x · f (x)dx.
−∞

Example
A (discrete) random variable is distributed with p(x) = x 2 /c, x = 1, 2, 3, 4.
What is the mean value?
What if the random variable is continuous in [1, 4]?

Answer: Z4 Z4 4
4 4
x3 100 x3 x4 63.75
x ·f (x) = = = .
X X
x · p(x) = = . c 4c c
c c x=1 x=1
1
x=1 x=1

4
P R4
From p(x) = 1, we get that c = 30. From f (x) = 1, we get that c = 21.
x=1 x=1

8 / 16 Chrysafis Vogiatzis Expectations

How to calculate means
P
Discrete random variables: E [X ] = x · p(x).
x∈S

+∞
R
Continuous random variables: E [X ] = x · f (x)dx.
−∞

Example
A (discrete) random variable is distributed with p(x) = x 2 /c, x = 1, 2, 3, 4.
What is the mean value?
What if the random variable is continuous in [1, 4]?

Answer: Z4 Z4 4
4 4
x3 100 x3 x4 63.75
x ·f (x) = = = .
X X
x · p(x) = = . c 4c c
c c x=1 x=1
1
x=1 x=1

4
P R4
From p(x) = 1, we get that c = 30. From f (x) = 1, we get that c = 21.
x=1 x=1

8 / 16 Chrysafis Vogiatzis Expectations

Means of discrete random variables

Bernoulli with probability p (assume failure=0 & success=1):

E [X ] = 1 · p + 0 · (1 − p) = p.

Binomial with parameters p and n:

E [X ] = n · p.
Example
Students accepted in a certificate program graduate with probability
p = 0.75. This year, the certificate program has accepted 300 students. How
many are expected to successfully finish the program?
Answer: Binomial with n = 300, p = 0.75, hence µ = n · p = 225 students.

9 / 16 Chrysafis Vogiatzis Expectations

Means of discrete random variables

Bernoulli with probability p (assume failure=0 & success=1):

E [X ] = 1 · p + 0 · (1 − p) = p.

Binomial with parameters p and n:

E [X ] = n · p.
Example
Students accepted in a certificate program graduate with probability
p = 0.75. This year, the certificate program has accepted 300 students. How
many are expected to successfully finish the program?
Answer: Binomial with n = 300, p = 0.75, hence µ = n · p = 225 students.

9 / 16 Chrysafis Vogiatzis Expectations

Means of discrete random variables

Bernoulli with probability p (assume failure=0 & success=1):

E [X ] = 1 · p + 0 · (1 − p) = p.

Binomial with parameters p and n:

E [X ] = n · p.
Example
Students accepted in a certificate program graduate with probability
p = 0.75. This year, the certificate program has accepted 300 students. How
many are expected to successfully finish the program?
Answer: Binomial with n = 300, p = 0.75, hence µ = n · p = 225 students.

9 / 16 Chrysafis Vogiatzis Expectations

Derivation of mean for binomial
From the definition of expectations for discrete random variables:

n
!
X X n x
E [X ] = x · p(x) = x p (1 − p)n−x =
x
x
x=0
n
X n!
= x px (1 − p)n−x =
x! · (n − x)!
x=0
n
X n · (n − 1)!
= p · px−1 · (1 − p)n−x =
(x − 1)! · (n − x)!
x=0
n
!
X n − 1 x−1
=np p (1 − p)n−x =
x −1
x=1
n−1
!
X n−1 k
=np p (1 − p)n−k −1 = (k = x − 1)
k
k =0
m
!
X m k
=np p (1 − p)m−k = (m = n − 1)
k
k =0

=np.
10 / 16 Chrysafis Vogiatzis Expectations
Means of discrete random variables
Geometric with parameter p:
1
E [X ] = .
p

Example
A kid learning basketball is shooting free throws with a probability of scoring
equal to 25%. What are the expected free throws the kid has to attempt until
scoring for the first time?
1
Answer: Geometric with p = 0.25, hence µ = p
= 4 free throws.
Hypergeometric with parameters N, K , n:
K
E [X ] = n .
N

Example
A trick-or-treat package contains 100 items, 20 of them are chocolate bars. A
kid picks 10 items at random; how many chocolate bars should they expect?
Answer: Hypergeometric with N = 100, K = 20, n = 10, hence µ = n KN = 2
chocolate bars.
11 / 16 Chrysafis Vogiatzis Expectations
Means of discrete random variables
Geometric with parameter p:
1
E [X ] = .
p

Example
A kid learning basketball is shooting free throws with a probability of scoring
equal to 25%. What are the expected free throws the kid has to attempt until
scoring for the first time?
1
Answer: Geometric with p = 0.25, hence µ = p
= 4 free throws.
Hypergeometric with parameters N, K , n:
K
E [X ] = n .
N

Example
A trick-or-treat package contains 100 items, 20 of them are chocolate bars. A
kid picks 10 items at random; how many chocolate bars should they expect?
Answer: Hypergeometric with N = 100, K = 20, n = 10, hence µ = n KN = 2
chocolate bars.
11 / 16 Chrysafis Vogiatzis Expectations
Means of discrete random variables
Geometric with parameter p:
1
E [X ] = .
p

Example
A kid learning basketball is shooting free throws with a probability of scoring
equal to 25%. What are the expected free throws the kid has to attempt until
scoring for the first time?
1
Answer: Geometric with p = 0.25, hence µ = p
= 4 free throws.
Hypergeometric with parameters N, K , n:
K
E [X ] = n .
N

Example
A trick-or-treat package contains 100 items, 20 of them are chocolate bars. A
kid picks 10 items at random; how many chocolate bars should they expect?
Answer: Hypergeometric with N = 100, K = 20, n = 10, hence µ = n KN = 2
chocolate bars.
11 / 16 Chrysafis Vogiatzis Expectations
Means of discrete random variables
Geometric with parameter p:
1
E [X ] = .
p

Example
A kid learning basketball is shooting free throws with a probability of scoring
equal to 25%. What are the expected free throws the kid has to attempt until
scoring for the first time?
1
Answer: Geometric with p = 0.25, hence µ = p
= 4 free throws.
Hypergeometric with parameters N, K , n:
K
E [X ] = n .
N

Example
A trick-or-treat package contains 100 items, 20 of them are chocolate bars. A
kid picks 10 items at random; how many chocolate bars should they expect?
Answer: Hypergeometric with N = 100, K = 20, n = 10, hence µ = n KN = 2
chocolate bars.
11 / 16 Chrysafis Vogiatzis Expectations
Means of discrete random variables
Geometric with parameter p:
1
E [X ] = .
p

Example
A kid learning basketball is shooting free throws with a probability of scoring
equal to 25%. What are the expected free throws the kid has to attempt until
scoring for the first time?
1
Answer: Geometric with p = 0.25, hence µ = p
= 4 free throws.
Hypergeometric with parameters N, K , n:
K
E [X ] = n .
N

Example
A trick-or-treat package contains 100 items, 20 of them are chocolate bars. A
kid picks 10 items at random; how many chocolate bars should they expect?
Answer: Hypergeometric with N = 100, K = 20, n = 10, hence µ = n KN = 2
chocolate bars.
11 / 16 Chrysafis Vogiatzis Expectations
Means of discrete random variables
Geometric with parameter p:
1
E [X ] = .
p

Example
A kid learning basketball is shooting free throws with a probability of scoring
equal to 25%. What are the expected free throws the kid has to attempt until
scoring for the first time?
1
Answer: Geometric with p = 0.25, hence µ = p
= 4 free throws.
Hypergeometric with parameters N, K , n:
K
E [X ] = n .
N

Example
A trick-or-treat package contains 100 items, 20 of them are chocolate bars. A
kid picks 10 items at random; how many chocolate bars should they expect?
Answer: Hypergeometric with N = 100, K = 20, n = 10, hence µ = n KN = 2
chocolate bars.
11 / 16 Chrysafis Vogiatzis Expectations
Means of discrete random variables
Geometric with parameter p:
1
E [X ] = .
p

Example
A kid learning basketball is shooting free throws with a probability of scoring
equal to 25%. What are the expected free throws the kid has to attempt until
scoring for the first time?
1
Answer: Geometric with p = 0.25, hence µ = p
= 4 free throws.
Hypergeometric with parameters N, K , n:
K
E [X ] = n .
N

Example
A trick-or-treat package contains 100 items, 20 of them are chocolate bars. A
kid picks 10 items at random; how many chocolate bars should they expect?
Answer: Hypergeometric with N = 100, K = 20, n = 10, hence µ = n KN = 2
chocolate bars.
11 / 16 Chrysafis Vogiatzis Expectations
 Poisson with parameter λ:

E [X ] = λ.

Note: when interested in finding the expected number of events given a

rate λ during a period t, we can find that as λ · t .
Example
A transportation engineer has installed a sensor to measure the number of
vehicles passing through an intersection. The number of vehicles is Poisson
distributed with rate λ = 60/hour . What is the expected number of vehicles
in 1 hour? What is the expected number of vehicles in 3 hours?
Answer: Poisson with rate λ = 60/hour , hence µ = λ = 60 vehicles. When
t = 3 hours, we should expect µ = λ · t = 180 vehicles.

12 / 16 Chrysafis Vogiatzis Expectations

 Poisson with parameter λ:

E [X ] = λ.

Note: when interested in finding the expected number of events given a

rate λ during a period t, we can find that as λ · t .
Example
A transportation engineer has installed a sensor to measure the number of
vehicles passing through an intersection. The number of vehicles is Poisson
distributed with rate λ = 60/hour . What is the expected number of vehicles
in 1 hour? What is the expected number of vehicles in 3 hours?
Answer: Poisson with rate λ = 60/hour , hence µ = λ = 60 vehicles. When
t = 3 hours, we should expect µ = λ · t = 180 vehicles.

12 / 16 Chrysafis Vogiatzis Expectations

Means of continuous random variables

Uniform between α and β:

α+β
E [X ] = .
2

Example
If the next bus arrives uniformly in the next 10 minutes, then the next bus is
expected to arrive in E [X ] = 5 minutes.

Normal with parameters µ, σ 2 :

E [X ] = µ.

Example
If grades are normally distributed with N (80, 12), then the expected grade of
a student in the class is E [X ] = 80.

13 / 16 Chrysafis Vogiatzis Expectations

Means of continuous random variables

Uniform between α and β:

α+β
E [X ] = .
2

Example
If the next bus arrives uniformly in the next 10 minutes, then the next bus is
expected to arrive in E [X ] = 5 minutes.

Normal with parameters µ, σ 2 :

E [X ] = µ.

Example
If grades are normally distributed with N (80, 12), then the expected grade of
a student in the class is E [X ] = 80.

13 / 16 Chrysafis Vogiatzis Expectations

Means of continuous random variables

Uniform between α and β:

α+β
E [X ] = .
2

Example
If the next bus arrives uniformly in the next 10 minutes, then the next bus is
expected to arrive in E [X ] = 5 minutes.

Normal with parameters µ, σ 2 :

E [X ] = µ.

Example
If grades are normally distributed with N (80, 12), then the expected grade of
a student in the class is E [X ] = 80.

13 / 16 Chrysafis Vogiatzis Expectations

 Exponential with rate λ:
1
E [X ] = .
λ

Example
If cars pass through an intersection with rate λ = 60/hour , then the next car
will pass in E [X ] = λ1 = 1 minute.

Gamma/Erlang with parameters λ and k :

k
E [X ] = .
λ

Example
If cars pass through an intersection with rate λ = 60/hour , then the
k = 30-th car is expected to pass in E [X ] = λk = 60/hour
30
= 0.5 hours.

14 / 16 Chrysafis Vogiatzis Expectations

 Exponential with rate λ:
1
E [X ] = .
λ

Example
If cars pass through an intersection with rate λ = 60/hour , then the next car
will pass in E [X ] = λ1 = 1 minute.

Gamma/Erlang with parameters λ and k :

k
E [X ] = .
λ

Example
If cars pass through an intersection with rate λ = 60/hour , then the
k = 30-th car is expected to pass in E [X ] = λk = 60/hour
30
= 0.5 hours.

14 / 16 Chrysafis Vogiatzis Expectations

 Exponential with rate λ:
1
E [X ] = .
λ

Example
If cars pass through an intersection with rate λ = 60/hour , then the next car
will pass in E [X ] = λ1 = 1 minute.

Gamma/Erlang with parameters λ and k :

k
E [X ] = .
λ

Example
If cars pass through an intersection with rate λ = 60/hour , then the
k = 30-th car is expected to pass in E [X ] = λk = 60/hour
30
= 0.5 hours.

14 / 16 Chrysafis Vogiatzis Expectations

Properties

Let α, β be real numbers and X , Y random variables. Then:

1. E [α] = α.
2. E [α · X ] = α · E [X ].
3. E [X + Y ] = E [X ] + E [Y ] .
 n
 n
P P
Generalizes to E Xi = E [Xi ].
i=1 i=1
 n
 n
P P
4. E αi · Xi = αi · E [Xi ]
i=1 i=1
5. E [a · X + b] = a · E [X ] + b.

Expected values of functions of random variables (g(X )):

P
discrete: E [g(X )] = g(x) · p(x).
x:p(x)>0
+∞
R
continuous: E [g(X )] = g(x) · f (x)dx.
−∞

15 / 16 Chrysafis Vogiatzis Expectations

Examples
Example
A company makes $2,000 if they sell 4 units, $1,800 if they sell 3 units,
$1,200 if they sell 2 units, lose $1,000 if they sell 1 unit, and lose $3,000 if
they sell no units. Each event from 0 to 4 customers is equally probable. How
much should they expect to make?
4
P
Answer: E [g(X )] = g(x) · p(x) =
x=0
1 1 1 1 1
2000 · 5
+ 1800 · 5
+ 1200 · 5
− 1000 · 5
− 3000 · 5
= $1000.

Example
Let X be a continuous random variable measuring the current (in
milliamperes, mA) in a wire with pdf f (x) = 0.05, for 0 ≤ x ≤ 20. The heat
produced from the current is given by the function g(x) = 10 · x (with x in
milliamperes). What is the mean heat produced by the current?

Answer:
R20 R20 R20
E [g(X )] = g(x) · f (x)dx = 10 · x · 0.05dx = 0.5xdx = 100.
x=0 x=0 x=0

16 / 16 Chrysafis Vogiatzis Expectations

Examples
Example
A company makes $2,000 if they sell 4 units, $1,800 if they sell 3 units,
$1,200 if they sell 2 units, lose $1,000 if they sell 1 unit, and lose $3,000 if
they sell no units. Each event from 0 to 4 customers is equally probable. How
much should they expect to make?
4
P
Answer: E [g(X )] = g(x) · p(x) =
x=0
1 1 1 1 1
2000 · 5
+ 1800 · 5
+ 1200 · 5
− 1000 · 5
− 3000 · 5
= $1000.

Example
Let X be a continuous random variable measuring the current (in
milliamperes, mA) in a wire with pdf f (x) = 0.05, for 0 ≤ x ≤ 20. The heat
produced from the current is given by the function g(x) = 10 · x (with x in
milliamperes). What is the mean heat produced by the current?

Answer:
R20 R20 R20
E [g(X )] = g(x) · f (x)dx = 10 · x · 0.05dx = 0.5xdx = 100.
x=0 x=0 x=0

16 / 16 Chrysafis Vogiatzis Expectations

Examples
Example
A company makes $2,000 if they sell 4 units, $1,800 if they sell 3 units,
$1,200 if they sell 2 units, lose $1,000 if they sell 1 unit, and lose $3,000 if
they sell no units. Each event from 0 to 4 customers is equally probable. How
much should they expect to make?
4
P
Answer: E [g(X )] = g(x) · p(x) =
x=0
1 1 1 1 1
2000 · 5
+ 1800 · 5
+ 1200 · 5
− 1000 · 5
− 3000 · 5
= $1000.

Example
Let X be a continuous random variable measuring the current (in
milliamperes, mA) in a wire with pdf f (x) = 0.05, for 0 ≤ x ≤ 20. The heat
produced from the current is given by the function g(x) = 10 · x (with x in
milliamperes). What is the mean heat produced by the current?

Answer:
R20 R20 R20
E [g(X )] = g(x) · f (x)dx = 10 · x · 0.05dx = 0.5xdx = 100.
x=0 x=0 x=0

16 / 16 Chrysafis Vogiatzis Expectations

Examples
Example
A company makes $2,000 if they sell 4 units, $1,800 if they sell 3 units,
$1,200 if they sell 2 units, lose $1,000 if they sell 1 unit, and lose $3,000 if
they sell no units. Each event from 0 to 4 customers is equally probable. How
much should they expect to make?
4
P
Answer: E [g(X )] = g(x) · p(x) =
x=0
1 1 1 1 1
2000 · 5
+ 1800 · 5
+ 1200 · 5
− 1000 · 5
− 3000 · 5
= $1000.

Example
Let X be a continuous random variable measuring the current (in
milliamperes, mA) in a wire with pdf f (x) = 0.05, for 0 ≤ x ≤ 20. The heat
produced from the current is given by the function g(x) = 10 · x (with x in
milliamperes). What is the mean heat produced by the current?

Answer:
R20 R20 R20
E [g(X )] = g(x) · f (x)dx = 10 · x · 0.05dx = 0.5xdx = 100.
x=0 x=0 x=0

16 / 16 Chrysafis Vogiatzis Expectations