Poisson Distribution
Poisson Distribution
1
Examples:
1. X = number of field mice per acre (𝑡 = 1 acre)
2. X= number of typing errors per page (𝑡 = 1
page)
3. X=number of telephone calls received every day
(𝑡 = 1 day)
4. X=number of telephone calls received every 5
days (𝑡 = 5 days)
2
• Let λ be the average (mean) number of
outcomes per unit time or unit region (𝑡
= 1).
• The average (mean) number of outcomes
(mean of X) in the time interval or region t
is:
𝜇 = 𝜆𝑡
3
The random variable X is called a Poisson
random variable with parameter 𝜇 (𝜇 = 𝜆𝑡), and
we write 𝑋~𝑃𝑜𝑖𝑠𝑠𝑜𝑛(𝜇) , if its probability
distribution is given by:
4
Theorem
The mean and the variance of the Poisson
distribution 𝑃𝑜𝑖𝑠𝑠𝑜𝑛(𝑥; 𝜇) are:
𝜇 = 𝜆. 𝑡
2
𝜎 = 𝜇 = 𝜆. 𝑡
5
Note:
• λ is the average (mean) of the distribution in the
unit time (𝑡 = 1).
• If X=The number of calls received in a month
(unit time 𝑡 = 1 month) and 𝑋~𝑃𝑜𝑖𝑠𝑠𝑜𝑛(𝜆), then:
(i) Y = number of calls received in a year.
𝑌 ~ 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 (𝜇); 𝜇 = 12𝜆 (𝑡 = 12)
(ii) W = number of calls received in a day.
𝑊 ~ 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 (𝜇); 𝜇 = 𝜆/30 (𝑡 = 1/30)
6
Example:
Suppose that the number of typing errors per page has
a Poisson distribution with average 6 typing errors.
(1) What is the probability that in a given page:
(i) The number of typing errors will be 7?
(ii) The number of typing errors will be at least 2?
(2) What is the probability that in 2 pages there will be
10 typing errors?
(3) What is the probability that in a half page there will
be no typing errors?
7
Solution:
(1) X = number of typing errors per page.
𝑋 ~ 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 (6) (𝑡 = 1, 𝜆 = 6, 𝜇 = 𝜆𝑡 = 6)
8
9
(2) X = number of typing errors in 2 pages
𝑋 ~ 𝑃𝑜𝑖𝑠𝑠𝑜𝑛(12) (𝑡 = 2, 𝜆 = 6, 𝜇 = 𝜆𝑡 = 12)
10
(3) X = number of typing errors in a half page.
𝑋 ~ 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 (3) (𝑡 = 1/2, 𝜆 = 6, 𝜇 = 𝜆𝑡 = 6/2 = 3)
11
Poisson approximation for binomial distribution
Let X be a binomial random variable with
probability distribution 𝑏(𝑥; 𝑛, 𝑝) . If 𝑛 → ∞, 𝑝
→ 0, and 𝜇 = 𝑛𝑝 remains constant, then the
binomial distribution 𝑏(𝑥; 𝑛, 𝑝) can be
approximated by Poisson distribution 𝑝(𝑥; 𝜇).
• For large n and small p we have:
𝑏(𝑥; 𝑛, 𝑝) ≈ 𝑃𝑜𝑖𝑠𝑠𝑜𝑛(𝜇) (𝜇 = 𝑛𝑝)
12
Example
X = number of items producing bubbles in a
random sample of 8000 items
𝑛 = 8000 and 𝑝 = 1/1000 = 0.001
𝑋 ~ 𝑏(𝑥; 8000, 0.001)
The exact probability is:
13
The approximated probability using Poisson
approximation:
𝑛 = 8000 (n is large, 𝑖. 𝑒. , 𝑛 → ∞)
𝑝 = 0.001 (𝑝 is small, 𝑖. 𝑒. 𝑝 → 0)
𝜇 = 𝑛𝑝 = 8000(0.001) = 8
𝑋 ≈ 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 (8)
14
Chapter 6
Some Continuous Probability
Distributions:
Continuous Uniform distribution
(Rectangular Distribution)
16
17
Theorem
The mean and the variance of the continuous
uniform distribution on the interval [A, B] are:
18
Example
Suppose that, for a certain company, the
conference time, X, has a uniform distribution on
the interval [0,4] (hours).
(a) What is the probability density function of X?
(b) What is the probability that any conference
lasts at least 3 hours?
19
Solution:
20
Normal Distribution
The normal distribution is one of the most
important continuous distributions.
Many measurable characteristics are normally or
approximately normally distributed, such as,
height and weight.
The graph of the probability density function
(pdf) of a normal distribution, called the normal
curve, is a bell-shaped curve.
21
22
The normal distribution characteristics
1) −∞ < 𝑋 < ∞
2) The density function of 𝑋 , 𝑓(𝑥), has a bell-shaped curve
𝑚𝑒𝑎𝑛 = 𝜇
𝑉𝑎𝑟𝑖𝑎𝑛𝑣𝑒 = 𝜎 2
−∞ 𝜇 +∞
3) The highest point of the curve of 𝑓(𝑥) at the mean 𝜇
The curve of 𝑓(𝑥) is symmetric about the mean 𝜇.
𝜇=mean = mode = median
𝜇1 = 𝜇2 < 𝜇3
𝜎1 > 𝜎2 = 𝜎3
The pdf of 𝑋 ~ 𝑁𝑜𝑟𝑚𝑎𝑙(𝜇, 𝜎) is given by:
26
Some properties of the normal curve 𝒇(𝒙)
of 𝑵(𝝁, 𝝈):
1. 𝑓(𝑥) is symmetric about the mean μ.
2. 𝑓(𝑥) has two points of inflection at 𝑥 = 𝜇 ± 𝜎.
3. The total area under the curve of 𝑓(𝑥) =1.
4. The highest point of the curve of 𝑓(𝑥) at the
mean 𝜇.
27
Areas Under the Normal Curve
The Standard Normal Distribution:
• The normal distribution with mean μ=0 and variance σ2=1
is called the standard normal distribution and is denoted by
Normal(0,1) or N(0,1). If the random variable Z has the
standard normal distribution, we write Z~Normal(0,1) or
Z~N(0,1).
28
• The pdf of Z~N(0,1) is given by:
29
• The standard normal distribution, Z~N(0,1), is
very important because probabilities of any normal
distribution can be calculated from the
probabilities of the standard normal distribution.
• Probabilities of the standard normal distribution
Z~N(0,1) of the form P(Z≤a) are tabulated.
30
31
• We can transfer any normal distribution
X~N(μ,σ) to the standard normal
distribution, Z~N(0,1) by using the
following result.
32
(ii) Probability of an interval event is given by the area
under the curve of 𝑓(𝑥) and above that interval.
Note: Probabilities of Z~N(0,1):
If X is continuous random variable then:
Finding 𝑃(𝑍 ≤ 𝑧) from the table
consider that the value of 𝑧 is rounded to 2 decimal places as 𝑧
= 𝑎. 𝑏𝑐
𝑎. 𝑏 𝑃(𝑍 ≤ 𝑎. 𝑏𝑐)
......
3.4
Example(1): 𝑍 ~ 𝑁(0,1)
1) 𝑃(𝑍 ≤ 1.50)
1.5 0.9332
......
0.9332
3.4
1.50
Example(2): 𝑍 ~ 𝑁(0,1)
2) 𝑃 𝑍 ≥ 0.98 = 1 − 𝑃 𝑍 ≤ 0.98 = 1 − 0.8365 = 0.1635
0.9 0.8365
......
0.1635
3.4
0.98
Example (3): 𝑍 ~ 𝑁(0,1)
3) 𝑃 −1.33 ≤ 𝑍 ≤ 2.42 = 𝑃 𝑍 ≤ 2.42 − 𝑃 𝑍 ≤ −1.33
= 0.9922 − 0.0918 = 0.9004
-1.3 0.0918
......
0.9004
2.4 0.9922
…..
-1.33 2.42
Example(4): 𝑍 ~ 𝑁(0,1)
4) 𝑃 𝑍 ≥ 0 = 𝑃 𝑍 ≤ 0 = 0.5
0.5 0.5
0
Notation
𝑃 𝑍 ≤ 𝑍𝐴 = 𝐴
𝐴 1−𝐴
𝑍𝐴
Example: Z ~ N(0,1)
Example: 𝑍 ~ 𝑁(0,1)