Scribd
Upload
46 views26 pages

Commonly Used Probability Distribution - SHORT

The document summarizes commonly used probability distributions for continuous and discrete random variables. For continuous variables, it describes the normal, lognormal, beta, and uniform distributions. For discrete variables, it discusses the discrete uniform, binomial, geometric, Poisson, and exponential distributions. It provides the probability mass/density functions and examples for some of the distributions.

Uploaded by

Sharif Md
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
46 views26 pages

Commonly Used Probability Distribution - SHORT

The document summarizes commonly used probability distributions for continuous and discrete random variables. For continuous variables, it describes the normal, lognormal, beta, and uniform distributions. For discrete variables, it discusses the discrete uniform, binomial, geometric, Poisson, and exponential distributions. It provides the probability mass/density functions and examples for some of the distributions.

Uploaded by

Sharif Md
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 26

MATH 2207

Commonly Used Probability Distribution


Reference Book: Probability, Reliability, and Statistical
Methods in Engineering Design
---Achintya Haldar, Sankaran Mahadevan

Chapter 4 1
Commonly Used Probability Distributions
• For Continuous Random Variable:
• Normal or Gaussian Distribution
One of the most commonly used distributions in engineering problems
It is applicable for any value of a random variable from −∞ to +∞.
• Lognormal Distribution
Suitable for a random variable cannot have negative values due to the physical
aspects of the problem e.g. Young’s modulus, weight, mass etc.
• Beta Distribution
It can be used when a random variable is known to be bounded by two limits,
a and b; e.g. CT marks varies within 0 and 10.
• Uniform Distribution
Special case of Beta Distribution.
Commonly Used Probability Distributions
• For Discrete Random Variable:
• Discrete Uniform Distribution
Used when outcomes are equally likely e.g. when a fair die is tossed, each element
of the sample space S = {1,2,3,4,5,6} occurs with probability 1/6.
• Binomial Distribution
Used when only two outcomes (success or failure) are possible, representing the
behavior of a discrete random variable and if the events satisfy the additional
requirements of a Bernoulli sequence.
Commonly Used Probability Distributions
• For Discrete Random Variable:
• Geometric Distribution
Suitable for the first occurrence time of an event e.g. for calculating the
probability of the failure of the structure in the 10th year.
• Poisson Distribution
Suitable for designing occurrence of events at any point in time or space.
• Exponential Distribution
The time between two consecutive occurrences of events (strong earthquakes),
can be modeled by an exponential distribution.
Normal or Gaussian Distribution
For random variable 𝑿 with mean 𝝁𝑿 and standard deviation 𝝈𝑿
• Probability Density Function (PDF):

• Cumulative Distribution Function (CDF):

The normal distribution is widely used and is denoted as 𝑵(𝝁, 𝝈).


Normal or Gaussian Distribution
The PDF of a normal distribution with a mean of 100 and a standard
deviation of 10 is as follows:
Features of the PDF:
• The distribution is
symmetric about the
mean.
• The mean, median,
and modal values are
identical.
Normal or Gaussian Distribution
The CDF of a normal distribution with a mean of 100 and a standard
deviation of 10 is as follows:
Normal or Gaussian Distribution
It is not simple to estimate the probability by integrating the PDF equation.
The problem can be addressed by transforming the original random variable,
X into a standard normal variable, S with zero mean (𝝁𝑺 = 𝟎) and unit
standard deviation (𝝈𝑺 = 𝟏), as

Transformed PDF and CDF are:

The standard normal distribution is denoted as 𝑵(𝟎, 𝟏) and its CDF is denoted as ϕ 𝒔 ,
that is, ϕ 𝑠 = 𝐹𝑆 (𝑠).
Normal or Gaussian Distribution
The CDF of the standard normal distribution is widely available in
tabulated form in Appendix-1 of the reference book:
Normal or Gaussian Distribution
The table in Appendix 1 is only valid
for positive values of the standard
normal random variable.
Since the normal distribution is
perfectly symmetrical, it can be
shown that:

or, when p < 0.5:


Normal or Gaussian Distribution
Example 4.1
Let X is a random variable following distribution
𝑵(𝟏𝟎𝟎, 𝟏𝟎). What is the probability that X will be below 99?
Solution:
𝑋 − 𝜇𝑋 99 − 100
𝑆= = = −0.1
𝜎𝑋 10
Normal or Gaussian Distribution
• From Appendix 1, we find that ϕ 0 = 0.5.
Therefore, CDF of the standard normal distribution evaluated at a
negative value of S will be less than 0.5.
• If a random variable X is 𝑵(𝝁𝑿 , 𝝈𝑿 ). then its probability of having a
value between two limits a and b is:
Normal or Gaussian Distribution
Example 4.2
In Young's modulus case study assume that the randomness in E can be
described by a normal random variable. Its mean and standard deviation
were estimated as 29,576 ksi and 1,507 ksi, respectively.
i. Calculate the probability of E having a value between 28,000 and 29,500 ksi.
ii. Calculate the probability of the Young's modulus being less than 2900 ksi.
iii. Calculate the probability of the Young's modulus being more than 2900 ksi.
iv. Calculate the 10th percentile value for the Young's modulus.

Hint for (i):


Binomial Distribution
It is applicable when−
• Only two outcomes (success or failure, good or bad) are possible,
representing the behavior of a discrete random variable.
• If the events satisfy the additional requirements of a Bernoulli
sequence:
i. If they are statistically independent
ii. If the probability of occurrence or nonoccurrence of events remains
constant
Binomial Distribution
If the probability of occurrence of an event in each trial is p and the
probability of nonoccurrence is (1 - p), then the probability of x
occurrences out of a total of n trials can be described by the PMF of a
binomial distribution as:

Here,
p = the probability of occurrence in each trial
𝑛 𝑛!
= = binomial coefficient
𝑥 𝑥! 𝑛−𝑥 !
X = the number of occurrences, is a discrete random variable, since it
can only take integer values
Binomial Distribution
Example 4.6
Binomial Distribution
Extra Example
The probability that a certain kind of component will survive a shock
test is 3/4.
An experimenter picks 15 components for test.
Find the probability that −
i. Exactly 2 components will survive
ii. 3 to 5 components will survive
iii. At most 2 components will fail
Geometric Distribution
• The first occurrence time of an event is of great interest in
engineering.
For example: information on the first time the design wind speed
will be exceeded in an area or the first time a structure will be
damaged by earthquakes is important.
• If the events—
i. occur in a Bernoulli sequence
ii. “p” is the probability of occurrence in each trial
then the probability that the event will occur for the first time at the
ith trial—is given by the geometric distribution as
P(T = t) = p(1-p)t-1; t = 1,2,....
Geometric Distribution
Probability, P(T = t) = p(1-p)t-1

• Example: Suppose the probability of failure of a structure due to


earthquakes is estimated as 10-5 per year. The probability of the
failure of the structure in the 8th year can be calculated as

P(T = 8) = 10-5(1- 10-5)8-1 = 9.9993× 10-6


Poisson Distribution
• It is suitable for designing occurrence of events at any point in time
or space.
For example:
oDefects can occur at any location along the length of welds.
oA tornado can strike a structure at any time during its lifetime.
oAn accident can occur at any location along a highway.
Poisson Distribution
• Suppose that the mean occurrence rate of tornadoes at a location is
v times a year. Thus, over a period of t years, tornadoes will occur
an average of “vt” times.
• If the time period t is divided into n intervals, then the probability
𝒗𝒕
of tornado occurrence in each interval will be .
𝒏
• Modeling x occurrences in time t in a Bernoulli sequence as n
approaches infinity will lead to the Poisson distribution, which can
be expressed as the following PMF:
𝒗𝒕 𝒙 −𝒗𝒕
P(x occurrences in time t) = 𝒆
𝒙!
• Similar concept is applicable for space in substitution of time.
Poisson Distribution
Example 4.9

𝒗𝒕 𝒙 −𝒗𝒕
P(x occurrences in time t) = 𝒆
𝒙!
Poisson Distribution
Example 4.11
Poisson Distribution
Example 4.11
Exponential Distribution
• If events occur according to a Poisson process, then the time T
before the first occurrence of the event—implying no occurrence
(i.e. x = 0) in time t—can be represented by the exponential
distribution. It can be shown that
𝒗𝒕 𝒙 −𝒗𝒕 𝒗𝒕 𝟎 −𝒗𝒕
P(T > t) = 𝒆 = 𝒆 = 𝒆−𝒗𝒕
𝒙! 𝟎!

• Corresponding CDF of T: 𝑭𝑻 𝒕 = 𝑷 𝑻 ≤ 𝒕 = 𝟏 − 𝒆−𝒗𝒕


• Corresponding PDF of T : 𝒇𝑻 𝒕 = 𝒗𝒆−𝒗𝒕 , 𝒕 ≥ 𝟎
𝟏
• Mean value of T:
𝒗
Exponential Distribution
Example 4.12

You might also like