Lecture 07 - Probability Distributions

EEF 271E
Probability and Statistics

Fall 2023
Lecture 07
Probability distributions
Probability distributions
Lecture Notes 07 1
Probability Distributions
Discrete Probability Distribution
– Bernoulli Distribution
– Binomial Distribution
– Geometric Binomial Distribution
– Negative Binomial Distribution
– Hypergeometric Distribution
– Poisson Distribution
Lecture Notes 07 Probability distributions 2

Continuous Probability Distributions
• Continuous Uniform Distribution

• Normal Distribution
• The Standard Normal Distribution
– Normal Approximation to Binomial
Distribution
• Exponential Distribution
• Erlang Distribution
• Gamma Distribution
• Chi-Square Distribution
• Weibull Distribution

Discrete Probability
Distribution

Discrete Random Variables
Definition: X is a discrete random variable,
𝑿 = 𝒙𝟏 , 𝒙𝟐 , 𝒙𝟑 , … is a set
if its range is countable (where the range means the
individual elements of the set.
𝑹𝑿 = 𝒙𝟏 , 𝒙𝟐 , 𝒙𝟑 , … .
We show the values in the range by lower case
letters such as 𝒙𝟏 , 𝒙𝟐 , 𝒙𝟑 , … etc.

Discrete Random Variables
𝑿 = 𝒙𝟏 , 𝒙𝟐 , 𝒙𝟑 , … is a set
if its range is countable (where the range means the
individual elements of the set.
𝑹𝑿 = 𝒙𝟏 , 𝒙𝟐 , 𝒙𝟑 , … .
We show the values in the range by lower case
letters such as 𝒙𝟏 , 𝒙𝟐 , 𝒙𝟑 , … etc.

Probability Mass Function – Discrete RV’s

𝑹𝒂𝒏𝒈𝒆 𝑿 = 𝑹𝑿 = 𝒙𝟏 , 𝒙𝟐 , 𝒙𝟑 , … .
The function:
𝑷𝑿 𝒙𝒌 = 𝑷 𝑿 = 𝒙𝒌 , 𝒇𝒐𝒓 𝒌 = 𝟏, 𝟐, 𝟑, 𝟑, … ,
is called the probability mass function (PMF) of X.

Example 1. Toss a fair coin twice, X = # of heads. Find the

range of X, RX as well as its probability mass function 𝑷𝑿 𝒙𝒌 .
If we repeat the experiment over and over and plot the
histogram, it will look like
The PMF in example 1
Random Variables 8
Lecture Notes 07
Example 2: X = # of rolls of a die until the first 6

appears.
Find the range of X, RX as well as its probability mass
function 𝑷𝑿 𝒙𝒌 .

Theorem:
For a discrete random variable with PMF 𝑷𝑿 𝒙 and Range
𝑹 𝑿 = 𝛀𝑿 = 𝒙 𝟏 , 𝒙 𝟐 , 𝒙 𝟑 , …
i. 𝟎 ≤ 𝑷𝑿 𝒙𝒌 ≤ 𝟏 𝒇𝒐𝒓 𝒂𝒍𝒍 𝒙𝒌 𝒙𝒌 ∈ 𝛀𝑿 .
ii. σ∞
𝒌=𝟏 𝑷 𝑿𝒌 = 𝟏
iii. 𝑨 ⊂ 𝛀𝑿 , 𝑷𝑿 𝑿 ∈ 𝑨 = 𝑷 𝑨
𝑷 𝑨 = ෍ 𝑷 𝑿 = 𝒙𝒌 = ෍ 𝑷𝑿 (𝒙𝒌 )
𝒙𝒌 ∈𝑨 𝒙𝒌 ∈𝑨

Independent Discrete Random Variables
Definition:
Consider n discrete random variables X1 , X2 , …Xn .
We say that X1 , X2 , …Xn are independent if
𝑷 𝑿 = 𝒙, 𝒀 = 𝒚 = 𝑷 𝑿 = 𝒙 𝑷 𝒀 = 𝒚 for all 𝒙, 𝒚
In general, if two random variables are independent,
then you can write
𝑷 𝑿 ∈ 𝑨, 𝒀 ∈ 𝑩 = 𝑷 𝑿 ∈ 𝑨 𝑷 𝒀 ∈ 𝑩 for all 𝑨 and𝑩

Independent Discrete Random Variables
Definition:
Consider two discrete random variables X and Y.
We say that X and Y are independent if
𝑷 𝑿𝟏 = 𝒙𝟏 , 𝑿𝟐 = 𝒙𝟐 , … , 𝑿𝒏 = 𝒙𝒏
= 𝑷 𝑿𝟏 = 𝒙𝟏 )𝑷 𝑿𝟐 = 𝒙𝟐 … 𝑷(𝑿𝒏 = 𝒙𝒏
for all 𝒙𝟏 , 𝒙𝟐 , … , 𝒙𝒏

Special Distributions

Bernoulli Distribution
Bernoulli random variables (RVs):
Example: Flip a coin {H,T}. Take an exam {Pass, Fail}.
𝑿 ~ 𝑩𝒆𝒓𝒏𝒐𝒖𝒍𝒍𝒊 (𝒑)
Probability Mass Function (PMF)
𝑝, 𝑓𝑜𝑟 𝑥 = 1
𝑃𝑋 𝑥 = ቊ
1 − 𝑝, 𝑓𝑜𝑟 𝑥 = 0
𝑃𝑋 0 = 1 − 𝑝, 𝑃𝑋 1 = 𝑝
IID Bernoulli RVs (e.g. Sequence of coin flips)
Where IID: Independent and Identical Distributed
Probability distributions 14
Lecture Notes 07
Binomial Distribution
❖ Repeated Trials:
❖ Repeat the same random experiment n times.
(Experiments are independent of each other)
❖ Number of times an event A happens among n
trials has Binomial distribution
(e.g. Number of heads in n coin tosses, number
of arrival in n times slots,…)
𝑿 ~ 𝑩𝒊𝒏𝒐𝒎𝒊𝒂𝒍 (𝒏, 𝒑)

𝑿 ~ 𝑩𝒊𝒏𝒐𝒎𝒊𝒂𝒍 (𝒏, 𝒑)
𝑛 𝑛 𝑛−𝑘
𝑃𝑋 𝑘 = 𝑝 1−𝑝 , k = 0, 1, 2, … , n
𝑘
𝐸 𝑋 = 𝑛𝑝, Var 𝑋 = 𝑛𝑝(1 − 𝑝)
Binomial is sum of n IID Bernoulli RVs.

Mean of Binomial Distribution
𝑛
𝑛 𝑘 𝑛−𝑘
𝐸 𝑋 = ෍𝑘 𝑝 1−𝑝
𝑘
𝑘=0
𝑛 𝑘
𝑛 𝑛 𝑝
= 1−𝑝 ෍𝑘
𝑘 1−𝑝
𝑘=0
𝑛−1
𝑛𝑛
𝑝 𝑝
= 1−𝑝 1+ = 𝑛𝑝
1−𝑝 1−𝑝
Note that
𝑛 𝑛
𝑛 𝑘 𝑛 𝑛 𝑘−1 𝑛−1
෍ 𝑥 = 1+𝑥 ⇒෍ 𝑥 =𝑛 1+𝑥
𝑘 𝑘
𝑘=0 𝑘=0

Binomial – Example (black and white dice)
A pair of dice (let say a black dice and a white dice)
tossed eight (8) times, and number of points on each
die added up.
(a) What is the probability that ‘’7 occurs exactly 4
times’’ ?
P({ 7 occurs exactly 4 times }) = ?
Ω = { (i,j): i, j =1,2,3,4,5,6 } and |Ω| = 36.

ωi,j = (i,j) i: # on black die, j: # on white die
E = “ the two dices together give 7”
E = {(1,6) (2,5), (3,4), (4, 3), (5,2), (6,1)}
|E| = 6
Lecture Notes 07
Example (black and white dice)
Lecture Notes 07
Example (black and white dice)
Lecture Notes 07
Example (black and White dice) :
Lecture Notes 07
Example (ball-bin) :
There are B bins, n ball are randomly

dropped into bins.
𝒑𝒊 : Probability that a ball goes to bin i.
𝐊 𝒊 : Number of balls in bin i after n drops.
𝑲𝒊 ~ 𝑩𝒊𝒏𝒐𝒎𝒊𝒂𝒍 (𝒏, 𝒑)
𝑖 = 1,2,3, … , 𝐵
Lecture Notes 07
Lemma: If 𝑿𝟏 , 𝑿𝟐 , … , 𝑿𝒏 are independent
Bernoulli random variable, then the random
variable 𝑋 defined by
𝑿 = 𝑿𝟏 , +𝑿𝟐 + ⋯ + 𝑿𝒏
is a Binomial(n, p) RV distribution.
Lecture Notes 07
Binomial - Example
n=4 p=0.2
n=10
n=20
n=40
Lecture Notes 07
DISCRETE PROBABILITY DISTRIBUTION
Geometric Binomial Distribution

• In a series of independent Binomial trials, with
constant probability p of a success, let the random
variable X denote the number of trials until the first
success. Then X has a geometric distribution with
parameter p and P(x) = (1 – p)x-1 p.
Random experiment: consider a coin with P(H)=p.
Toss the coin repeatedly until the first heads is observed.
𝑋 = The total number of coin tosses
𝑋 ~ 𝐺𝑒𝑜𝑚(𝑝)

Geometric Distribution
• Repeated trials: Number of trials some event occurs.
𝑋 ~ 𝐺𝑒𝑜𝑚(𝑝)
𝑅𝑋 = 𝑅𝑎𝑛𝑔𝑒 𝑋 = 1,2,3, … ∈ ℕ
𝑘−1
𝑃𝑋 𝑘 = 𝑃 𝑋 = 𝑘 = 1 − 𝑝 𝑝 𝑘 = 1,2,3, …
• The mean and the variance of X are
1 2
1−𝑝
𝐸 𝑋 =𝜇= , 𝑉𝑎𝑟 𝑋 = 𝜎 = 2
𝑝 𝑝
• The cumulative distribution function of X
𝑃𝑋 𝑋 > 𝑘 = 1 − 𝑝 𝑘
𝐹𝑋 𝑋 ≤ 𝑘 = 1 − 1 − 𝑝 𝑘

Negative Binomial Distribution
(Pascal Distribution)
• Is a series of independent Binomial trials, with constant
probability p of a success, let the random variable X
denote the number of trials until r successes occur.
Then X has a negative distribution with parameter p
and
r = 1, 2, 3, … , and
𝑥−1
𝑃(𝑥) = (1 − 𝑝)𝑥−𝑟 𝑝𝑟
𝑟−1
for x = r, r + 1, r + 2,….
Negative Binomial Distribution (Pascal Distribution)

Example. You flip a coin until you observe r heads.
𝑿 = Total number of coin toss
𝑹𝑿 = 𝒓, 𝒓 + 𝟏, 𝒓 + 𝟐, … ∈ ℕ
Find PMF.

Negative Binomial Distribution (Pascal Distribution)

𝐴 = 𝑋 = 𝑘 𝑜𝑟 𝐴 = 𝐵 ∩ 𝐶.
B is the event that we observe (r−1) heads in the first (𝑘−1) trials.
C is the event that we observe a heads in the kth (the last) trial.
𝑃 𝐴 = 𝑃𝑋 𝑘 = 𝑃(𝑋 = 𝑘)
𝑃 𝐴 = 𝑃 𝐵 ∩ 𝐶 = 𝑃 𝐵 𝑃 𝐶 , B and C are independent events.
Using binomial formula 𝐁𝐢𝐧𝐨𝐦𝐢𝐚𝐥 (𝐧 = 𝐤 − 𝟏, 𝐩)
𝑘 − 1 𝑟−1 [ 𝑘−1 − 𝑟−1 ]
𝑃 𝐵 = 𝑝 1−𝑝
𝑟−1
𝑘 − 1 𝑟−1
= 𝑝 1 − 𝑝 𝑘−𝑟
𝑟−1
Pascal Distribution (Negative Binomial Distribution)

Definition: A random variable 𝑿 said to be a Pascal random
variable with parameters r and p ( 𝑃 𝐻 = 𝑝), shown as
𝑋 ~ 𝑃𝑎𝑠𝑐𝑎𝑙(𝑟, 𝑝) if
𝑘 − 1 𝑟−1 𝑘−𝑟
𝑃𝑋 𝑟 = 𝑝 1−𝑝
𝑟−1
where 0 < 𝑝 < 1.

Hypergeometric Distribution
• A set of N objects contains
• K objects classified as success and
• N – K objects classified as failures
• A sample of size n objects is selected, at random
(without replacement) from the N objects, where K  N
and n  N.
• Let the random variable X denote the number of
successes in the sample. Then X has a
Hypergeometric distribution and

Hypergeometric Distribution
𝐾 𝑁−𝐾
𝑃(𝑥) = 𝑥 𝑛 − 𝑥 𝑥 = max{ 0, 𝑛 + 𝐾 − 𝑁}𝑡𝑜 min{ 𝐾, 𝑛}
𝑁
𝑛
The mean and the variance are
𝑁−𝑛
𝜇=𝐸 𝑋 = 𝑛𝑝𝑎𝑛𝑑𝜎 2 = 𝑛𝑝 1 − 𝑝
𝑁−1
𝑤ℎ𝑒𝑟𝑒 𝑝 = 𝐾/𝑁

Example (Hypergeometric Distribution)
Suppose that N = 12 resistor have just come off the
production line. Also assume that n = 3 are defective (have
serious defects).
Thus b = N-n = 12-3 = 9 are defect-free.
A sample of n = 5 are selected and tested to see if they are
defective.
Let X = the number defective resistors in the sample
(X:{0,1,2,3}).
a) Find the probability mass function of X and draw its
graph.
b) Find the expected value of the defective resistor for
this case (i.e., in the case with 12 resistors have
produced in a line).
Lecture Notes 07
Answer (Hypergeometric Distribution)
a) This is a hypergeometric distribution, so we have the
following probabilities:
𝑁 = 12, 𝑛 = 5, 𝐾 = 3, 𝑏 = 𝑁 − 𝐾 = 12 − 3 = 9
𝐾 𝑏 3 9
𝑥 𝑛−𝑥
𝑃 𝑋=𝑥 = = 𝑥 5−𝑥 , 𝑥 = 0, 1,2,3
𝑁 12
𝑛 5
3 9
𝑃 𝑋=0 = 0 5 − 0 = 1 × 126 = 0.159
12 792
5
3 9
𝑃 𝑋=1 = 1 5 − 1 = 3 × 126 = 0.477
12 792
5
Lecture Notes 07
Answer (Hypergeometric Distribution)
3 9
𝑃 𝑋=2 = 2 5 − 2 = 3 × 84 = 0.31818
12 792
5
3 9
𝑃 𝑋=3 = 3 5 − 3 = 1 × 36 = 0.045
12 792
5
P(X = xi)
xi P(X = xi)
0.5
0 0.159
1 0.477
2 0.381818
3 0.045
0 1 2 3 x
Lecture Notes 07
Poisson Distribution
Poisson distribution is another important

probability distribution of a discrete
random variable that has a large number
of applications. It is often used for events
that occur at constant rate.

Poisson Random Variable:

Poisson RVs are used to model
• Arrival of customers at a service facility
• Arrival of information request at a server
• Etc. (for events that occur at constant rate).
Counting the occurrence of certain events in an interval
of time or space.

Arrival of customers in an interval:

λ: the average number of arrivals in that interval


Used to model number of occurrences or arrivals

etc.
𝑿 ~ 𝑷𝒐𝒊𝒔𝒔𝒐𝒏 (𝝀)
𝑘
𝜆
𝑃𝑋 𝑘 = 𝑒 𝜆 , k = 0, 1, 2, …
𝑘!
𝐸 𝑋 = 𝜆, Var 𝑋 = 𝜆
Poisson Graphs
0.5
0.45 l=.5
0.4
0.35
l=1
0.3
0.25
l=4
0.2
0.15 l=10
0.1
0.05
0
0 5 10 15
Winter 2006 EE384x 41

Poisson approximation to Binomial Distribution.

𝜆 =𝑛∙𝑝
𝑘
−(𝑛𝑝)
(𝑛𝑝)
𝑃 𝑋 = 𝑘 = 𝑃𝑋 𝑘 = 𝑒
𝑘!

Poisson as limit of Binomial
• Poisson is the limit of Binomial(n,p) as
• Let
43

Poisson and Binomial
0.4
0.35 n=5,p=4/5
0.3 Poisson(4)
n=10,p=.4
0.25
0.2
n=20, p=.2
0.15
0.1
n=50,p=.08
0.05
0
0 1 2 3 4 5 6 7 8 9 10
45
Example:
The expected number of car accidents in an intercity highway in a
month is known as 4.5, (i.e., it has is known to be l = 4.5). The
number of accidents is known and to have a Poisson distribution.
a) Compute the probability of exactly no accident in a month
b) Compute the probability of exactly 1 accident in a month
c) Compute the probability of at most 2 accidents in a month
d) Compute the probability of at least 4 accidents in a month.
e) Compute the probability of at most 4 accidents in a month, if it
is known that there were at least 2 accidents. (Assume that the
number of accidents for a month in the city follows a Poisson
distribution)

Answer :
𝜆 𝑥 −𝜆
𝑃 𝑋=𝑥 =𝑝 𝑥 = 𝑒 , 𝑥 = 0,1,2,3, … , 8
𝑥!
4.5 0 −4.5
a) 𝑃 𝑋=0 =𝑝 0 = 𝑒 = 0.0011109
0!
4.5 1 −4.5
b) 𝑃 𝑋=1 =𝑝 1 = 𝑒 = 0.0499
1!
c) 𝑃 𝑋 ≤2 =𝑃 𝑋 =0 +𝑃 𝑋 =1 +𝑃 𝑋 =2
= 𝑝 0 + 𝑝 1 + 𝑝(2)
4.5 2 −4.5
𝑝 2 = 𝑒 = 0.112479
2!
𝑃 𝑋 ≤2 =𝑝 0 +𝑝 1 +𝑝 2
= 0.0011109 + 0.0499 + 0.112479 = 0.173578

Answer (continued):
d) 𝑃 𝑋 ≥ 4 = 1 − 𝑃 𝑋 ≤ 3 =
=1− 𝑝 0 +𝑝 1 +𝑝 2 +𝑝 3
4.5 3 −4.5
𝑝 3 = 𝑒 = 0.168718
3!
𝑃 𝑋 ≥ 4 = 1 − 𝑃 𝑋 ≤ 3 = 1 − 0.342296 = 0.657704
e)
𝑃 𝑋≤4 ∩ 𝑋≥2 𝑃 2≤𝑋≤4
𝑃 𝑋 ≤ 4|𝑋 ≥ 2 = =
𝑃 𝑋≥2 1−𝑃 𝑋 ≤1
𝑝 2 + 𝑝 3 + 𝑝(4)
=
1 − [𝑝 0 + 𝑝 1 + 𝑝 2 ]
4.5 4 −4.5
𝑝 4 = 𝑒 = 0.189808
4!

CONTINUOUS RANDOM VARIABLES AND
THEIR PROBABILITY DISTRIBUTIONS
Continuous Probability Distributions

• Continuous Uniform Distribution
• Normal Distribution
• The Standard Normal Distribution
– Normal Approximation to Binomial Distribution
• Exponential Distribution
• Erlang Distribution
• Gamma Distribution
• Chi-Square Distribution
• Weibull Distribution
Lecture notes 15 Probability distributions 49

Continuous Probability
Distributions

Continuous Random Variables
If 𝐑𝐚𝐧𝐠𝐞(𝑿) is not countable then 𝑿 is not a discrete

random variable.
Example: 𝐑𝐚𝐧𝐠𝐞 𝑿 = [𝒂, 𝒃]
Remember: 𝒂, 𝒃 = 𝒙 ∈ ℝ, 𝒂≤𝒙≤𝒃 ,
(𝒂, 𝒃] = 𝒙 ∈ ℝ, 𝒂<𝒙≤𝒃
Example.
T: Lifetime of a light bulb, 𝐑𝐚𝐧𝐠𝐞 𝑻 = [𝟎, ∞)
V: DC Voltage across a resistor, 𝐑𝐚𝐧𝐠𝐞 𝑽 = [𝟎, 𝑽𝒎𝒂𝒙 ]

If the CDF 𝑭𝑿 (𝒙) is a continuous function, then 𝑿 is said to be

a continuous random variable.
PMF is Not well-defined for continuous random variables.
Instead, we define probability density function (pdf), 𝒇𝑿 (𝒙).
Definition. A random variable 𝑿 having a CDF 𝑭𝑿 (𝒙)

that is a continuous function for all 𝒙 in ℝ is said to be a
continuous random variable.

For a discrete random variable: the rate of increase in the CDF

𝑭𝑿 𝒙 is characterized by the PMF 𝑷𝑿 𝒙 - that is, by the locations
and sizes of the jumps in the CDF.
What characterizes the rate of increase for a continuous function?

The derivative of the function

Continuous Uniform Distribution
Example.
Let 𝑎, 𝑏 be an interval in the real line (where 𝑎 and
𝑏 are real numbers with 𝑎 < 𝑏). Let 𝑋 be a number
chosen at random from that interval.
“Chosen at random” means: If 𝑎 ≤ 𝑥1 ≤ 𝑥2 ≤ 𝑏, then
𝑥2 − 𝑥1
𝑃 𝑋 ∈ [𝑥1 , 𝑥2 ] =
𝑏−𝑎

Let’s find the CDF
𝐚) 𝑭𝑿 𝒙 = 𝟎, for 𝒙 < 𝒂,
𝒙−𝒂
𝒃) 𝑭𝑿 𝒙 = , for 𝒙 ≤ 𝒂 ≤ 𝐛,
𝒃−𝒂
𝒄) 𝑭𝑿 𝒙 = 𝟏, for 𝒙 ≥ 𝐛,
In this case, 𝑿 is called a 𝑼𝒏𝒊𝒇𝒐𝒓𝒎 𝒂, 𝒃 random variable.

Probability Density Function (PDF)
Theorem.
𝑷 𝒂 < 𝑿 < 𝒃 = 𝑭𝑿 𝒃 − 𝑭𝑿 𝒂
𝒃 𝒂 𝒃
= න 𝒇𝑿 𝝉 𝒅𝝉 − න 𝒇𝑿 𝝉 𝒅𝝉 = න 𝒇𝑿 𝝉 𝒅𝝉
−∞ −∞ 𝒂
Where
𝑭𝑿 𝒙 is the CDF (Cumulative Distribution Function),
𝒇𝑿 𝒙 is the PDF (Probability Density Function)

Probability Density Function (PDF) of a continuous
uniform function
Example. Say that 𝑿 ~ 𝑼𝒏𝒊𝒇𝒐𝒓𝒎 𝒂, 𝒃

The mean and the variance of a Continuous Uniform Distribution
Probability density function Expected value (Mean value),E(X)
1 𝑎+𝑏
𝑓(𝑥) = ,𝑎 ≤ 𝑥 ≤ 𝑏 𝐸(𝑋) = 𝜇 =
𝑏−𝑎 2
Variance, Var(X)
f(x)
(𝑏 − 𝑎)2
𝑉𝑎𝑟(𝑋) = 𝜎 2 =
1 / (b-a) 12
a b x

CONTINUOUS RANDOM VARIABLES AND THEIR PROBABILITY DISTRIBUTIONS

EXAMPLE
Let the continuous variable x denote the current measured in a thin copper
wire in millimeter. Assume that the range of x is [0, 20mA], and assume that
the probability density function of x is f(x) = 0.05, 0 <= x <= 20.
What is the probability that a measurement of current is between 5 and 10
millimeter? What are the mean and the variance?
𝑏
• Solution 𝑃(𝑎 ≤ 𝑥 ≤ 𝑏) = න 𝑓(𝑥)𝑑𝑥

𝑎
10
𝑃(5 ≤ 𝑥 ≤ 10) = න 0.05𝑑𝑥 = 0.05(10 − 5) = 0.25

5
𝑎+𝑏
f(x) 𝜇 = 𝐸(𝑋) = = (0 + 20)/2 = 10
2
𝑎𝑛𝑑
0.05
2
(𝑏 − 𝑎)2
𝜎 = 𝑉(𝑋) = = 400/12
12
= 33.33
0 5 10 20 mA x

Normal Distribution
The normal distribution is the most important and

most widely used of all probability distributions.
A normal probability distribution or the normal

curve, when plotted, gives a bell-shaped curve
such that
The total area under the curve is 1.0.

The curve is symmetric about the mean.
The two tails of the curve extend indefinitely.

Normal Distribution

Normal Distribution
A normal distribution random variable has the

following probability density function
1 −(𝑥−𝜇)2
𝑓 𝑥 = 𝑒 2𝜎2 𝑓𝑜𝑟 − ∞ < 𝑥 < ∞
𝜎 2𝜋

Normal Distribution

Normal Distribution

Normal Distribution

The Standard Normal Distribution
The standard normal distribution is a special case of the normal
distribution.
The normal distribution with  = 0 and  = 1 is called standard
normal distribution.
Converting an x value to a z value
• A random variable that has a standard normal distribution is
called standard normal random variable, z, and it is equal to
𝑥−𝜇
𝑧=
𝜎
• A standard normal random variable, z, has the following

probability density function
• The mean and the variance for z are as follows


A standard normal random variable, z, has the
following probability density function
The mean and the variance for z are as follows
1 −𝑧 2
𝑓(𝑧) = 𝑒 2 𝑓𝑜𝑟 −∞<𝑥 <∞
2𝜋
The mean and the variance for z are as follows

E(z) = 0 and V(z) = 1

The cumulative distribution function of normal random variable is denoted as
z −u 2
1
( z ) = P( Z  z ) =
2 e
−
2 du
+ −u 2
1
( z ) =
2 e
−
2 du = 1
Table II provides cumulative probability values for , for a standard normal

random variable.

EXAMPLE:
Assume Z is a standard normal random variable. Table II provides probabilities of
the form P( Z  z ) . The use of Table II to find P( Z  1.5) is illustrated in following
figure Read down the z column to the row that equals 1.5. The probability is read
from the adjacent column, labeled 0.00, to be 0.93319.

EXAMPLE

EXAMPLE
The time it takes a cell to divide (called mitosis) is normally distributed with
an average time of one hour and a standard deviation of 5 minutes.
(a) What is the probability that a cell divides in less than 45 minutes?
(b) What is the probability that it takes a cell more than 65 minutes to divide?
(c) What is the time that it takes approximately 99% of all cells to complete
mitosis?

Normal Approximation to Binomial Distribution
If X is a binomial random variable, then
X − np
Z=
np(1 − p)
is approximately a standard normal random variable.
Usually, the normal distribution is used as an approximation to the binomial

distribution when np and n(1-p) are both greater than 5, that is, when
np > 5 and n(1-p) > 5

np > 5 and n(1-p) > 5
9
8
7
6
5
4
3
2
1
0
1 2 3 4 5 6 7 8 9

The area of each bar equals the binomial probability of x.

Notice that the area of bars can be approximated by areas
under the normal density function.

Exponential Distribution
The exponential distribution is often used in reliability studies as the model for the time
until failure of a device.
Examples
The lifetime of a semiconductor chip might be modeled as an exponential random

variable with a mean of 40,000 hours.
The lifetime of a device with failures caused by random shocks might be appropriately
modeled as an exponential random variable.
The random variable X that equals the distance between successive counts of a Poisson
process with mean l > 0 has an exponential distribution with parameter l. The
probability density function of X is
f ( x) = l e −lx , for 0  x  
1 1
 = E( X ) = and  2 = V ( X ) =
l l2

Exponential Distribution
Probability density function of an exponential random variable for

selected values of l

Erlang Distribution
An exponential random variable describes the length until the first

count is obtained in a Poisson process. A generalization of the
exponential distribution is the length unit r counts occur in a
Poisson process. The random variable that equals the interval
length until r counts occur in a Poisson process is an Erlang
random variable.
lr x r −1e − lx
f ( x) = , for x  0 and r = 1,2,...
(r − 1)!
r
 = E( X ) =
l
r
 2 = V (X ) =
l2

Gamma Distribution
The Erlang distribution is a special case of the gamma distribution. If the

parameter r of an Erlang random variable is not integer, but r > 0, then the
random variable has a gamma distribution.
lr x r −1e − lx
f ( x) = , for x  0
( r )

The gamma function ( x) =  x r −1e − x dx, for r  0
0
r r
 = E( X ) = and  2 = V ( X ) =
l l2

Chi-Square Distribution
A special case of the gamma distribution in which
l = ½ and r equals one of the values 0.5, 1, 1.5, 2, ….
This distribution is used extensively in interval

estimation and test of hypotheses.

Weibull Distribution
The Weibull distribution is often used to model the time until failure of many
different physical systems. The parameters in the distribution provide a great
deal of flexibility to model systems in which the number of failures increases
with time (bearing wear), decreases with time (some semiconductors), or
remains constant with time (failures caused by external shocks to the system).
The random variable X with probability density function
 −1
 x −( x /  ) 
f ( x) =   e , for x  0
  
has a Weibull distribution with scale parameter >0 and

shape parameter >0.

2
 1 2   1 
 =  1 +  and  2 =  2 (1 + ) −  2 1 + 
      

Weibull probability density functions for selected values of  and 

Lecture 07 - Probability Distributions

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lecture 07 - Probability Distributions

Uploaded by

Copyright:

Available Formats

EEF 271E

Probability and Statistics

Lecture Notes 07 Probability distributions 2

• Continuous Uniform Distribution

Lecture Notes 07 Probability distributions 3

Lecture Notes 07 Probability distributions 4

if its range is countable (where the range means the

individual elements of the set.

We show the values in the range by lower case

letters such as 𝒙𝟏 , 𝒙𝟐 , 𝒙𝟑 , … etc.

Lecture Notes 07 Probability distributions 5

if its range is countable (where the range means the

individual elements of the set.

We show the values in the range by lower case

letters such as 𝒙𝟏 , 𝒙𝟐 , 𝒙𝟑 , … etc.

Lecture Notes 07 Probability distributions 6

Definition: X is a discrete random variable,

is called the probability mass function (PMF) of X.

Lecture Notes 07 Probability distributions 7

Example 1. Toss a fair coin twice, X = # of heads. Find the

The PMF in example 1

Example 2: X = # of rolls of a die until the first 6

Lecture Notes 07 Probability distributions 9

Lecture Notes 07 Probability distributions 10

Lecture Notes 07 Probability distributions 11

Lecture Notes 07 Probability distributions 12

Lecture Notes 07 Probability distributions 13

Lecture Notes 07 Probability distributions 15

𝐸 𝑋 = 𝑛𝑝, Var 𝑋 = 𝑛𝑝(1 − 𝑝)

Binomial is sum of n IID Bernoulli RVs.

Lecture Notes 07 Probability distributions 16

Lecture Notes 07 Probability distributions 17

Ω = { (i,j): i, j =1,2,3,4,5,6 } and |Ω| = 36.

There are B bins, n ball are randomly

𝐊 𝒊 : Number of balls in bin i after n drops.

Geometric Binomial Distribution

Lecture Notes 07 Probability distributions 25

Lecture Notes 07 Probability distributions 26

Negative Binomial Distribution (Pascal Distribution)

Lecture Notes 07 Probability distributions 28

Negative Binomial Distribution (Pascal Distribution)

Pascal Distribution (Negative Binomial Distribution)

Lecture Notes 07 Probability distributions 30

Lecture Notes 07 Probability distributions 31

The mean and the variance are

Lecture Notes 07 Probability distributions 32

Poisson distribution is another important

Lecture Notes 07 Probability distributions 36

Poisson Random Variable:

Lecture Notes 07 Probability distributions 37

Arrival of customers in an interval:

Lecture Notes 07 Probability distributions 38

Lecture Notes 07 Probability distributions 39

Used to model number of occurrences or arrivals

Winter 2006 EE384x 41

Poisson approximation to Binomial Distribution.

Lecture Notes 07 Probability distributions 42

Lecture Notes 07 Probability distributions 44

Lecture Notes 07 Probability distributions 46

Lecture Notes 07 Probability distributions 47

Lecture Notes 07 Probability distributions 48

Continuous Probability Distributions