MODULE 5
OTHER DISCRETE PROBABILITY DISTRIBUTION
TLO 6: Solve some probability distributions for
continuous random variables
BINOMIAL DISTRIBUTION
• BERNOULLI PROCESS
- A random experiment that consists of repeated trials, each of which
has two possible outcomes that may be defined as SUCCESS or
FAILURE
- Each trial in the process is called a Bernoulli trial
- The number, x, of successes in an n Bernoulli trial is called a BINOMIAL
RANDOM VARIABLE
• the probability distribution is called a BINOMIAL DISTRIBUTION
BINOMIAL DISTRIBUTION
• BERNOULLI PROCESS
- Must satisfy the following properties:
- B - Binary Outcomes: Every trial must result in exactly one of two
outcomes. For example, a material passes or fails a stress test, or a weld is
sound or defective.
- I - Independent Trials: The outcome of one trial cannot affect the next. For
example, testing one block does not change the strength of the next
block.
- N - Number of Trials is Fixed: You must know exactly how many items you
are testing (n) before you start.
- S - Same Probability: The probability of "success" (p) remains constant for
every single trial.
BINOMIAL DISTRIBUTION
• SUMMARY:
• a Bernoulli trial can result in a success with probability 𝐩 and a
failure with probability 𝐪 = 𝟏 − 𝐩
• Then the probability distribution of the binomial random variable
X, number of successes in 𝒏 independent trials is
• 𝑷(𝑿 = 𝒙) = 𝒏𝑪𝒙 𝒑𝒙 𝒒𝒏 𝒙 , 𝒙 = 𝟎, 𝟏, 𝟐, … , 𝒏
BINOMIAL DISTRIBUTION
• 1. You are the QA/QC engineer on site. You are verifying a delivery of load-
bearing Concrete Hollow Blocks (CHB) intended for a major masonry wall
project. Based on the manufacturer's historical data, the probability that a
single CHB will fail the compressive strength test is 5 %. You randomly
select 10 blocks for testing. What is the probability that exactly 2 out of
the 10 blocks fail the test?
• Soln: n = 10, x = 2, p = 0.05, q = 0.95
• 𝑷(𝑿 = 𝒙) = 𝒏𝑪𝒙 𝒑𝒙 𝒒𝒏 𝒙
• P(X = 2) = (10C2) * (0.05^2) * (0.95^8)
• P(X = 2) = 0.0746
BINOMIAL DISTRIBUTION
• 2. The probability that a certain kind of component will survive a shock test
is 3/4. Find the probability that exactly 2 of the next 4 components tested
survive.
Soln: n = 4, x = 2, p = 0.75, q = 0.25
• 𝑷(𝑿 = 𝒙) = 𝒏𝑪𝒙 𝒑𝒙 𝒒𝒏 𝒙
• P(X = 2) = (4C2) * (0.75^2) * (0.25^2)
• P(X = 2) = 0.2109
BINOMIAL DISTRIBUTION
• 3. The probability that a patient recovers from a rare blood disease is 0.4. If 15
people are known to have contracted this disease, what is the probability that (a)
at least 10 survive, (b) from 3 to 8 survive, and (c) exactly 5 survive?
• Soln: n = 15 (total patients) p = 0.4 q = 0.6
• (a) at least 10 survive
• P(X ≥ 10) = P(X=10) + P(X=11) + P(X=12) + P(X=13) + P(X=14) + P(X=15)
• 𝑷(𝑿 = 𝒙) = 𝒏𝑪𝒙 𝒑𝒙 𝒒𝒏 𝒙
• P(X=10) = (15C10) * (0.4^10) * (0.6^5) = 0.0245
P(X=11) = (15C11) * (0.4^11) * (0.6^4) = 0.0074 P(X ≥ 10) = 0.0338
P(X=12) = (15C12) * (0.4^12) * (0.6^3) = 0.0016
P(X=13) = (15C13) * (0.4^13) * (0.6^2) = 0.0003
P(X=14) = (15C14) * (0.4^14) * (0.6^1) = 0.0000 (negligible)
P(X=15) = (15C15) * (0.4^15) * (0.6^0) = 0.0000 (negligible)
BINOMIAL DISTRIBUTION
• 3. The probability that a patient recovers from a rare blood disease is 0.4. If 15
people are known to have contracted this disease, what is the probability that (a)
at least 10 survive, (b) from 3 to 8 survive, and (c) exactly 5 survive?
• Soln: n = 15 (total patients) p = 0.4 q = 0.6
• (b) from 3 to 8 survive,
• P(3 ≤ X ≤ 8) = P(X=3) + P(X=4) + P(X=5) + P(X=6) + P(X=7) + P(X=8)
• 𝑷(𝑿 = 𝒙) = 𝒏𝑪𝒙 𝒑𝒙 𝒒𝒏 𝒙
• P(X=3) = (15C3) * (0.4^3) * (0.6^12) = 0.0634
P(X=4) = (15C4) * (0.4^4) * (0.6^11) = 0.1268
P(X=5) = (15C5) * (0.4^5) * (0.6^10) = 0.1859 P(3 ≤ X ≤ 8) = 0.8779
P(X=6) = (15C6) * (0.4^6) * (0.6^9) = 0.2066
P(X=7) = (15C7) * (0.4^7) * (0.6^8) = 0.1771
P(X=8) = (15C8) * (0.4^8) * (0.6^7) = 0.1181
BINOMIAL DISTRIBUTION
• 3. The probability that a patient recovers from a rare blood disease is 0.4. If
15 people are known to have contracted this disease, what is the
probability that (a) at least 10 survive, (b) from 3 to 8 survive, and (c) exactly
5 survive?
• Soln: n = 15 (total patients) p = 0.4 q = 0.6
• (c) exactly 5 survive
P(X = 5)
𝑷(𝑿 = 𝒙) = 𝒏𝑪𝒙 𝒑𝒙 𝒒𝒏 𝒙
• P(X=5) = (15C5) * (0.4^5) * (0.6^10) = 0.1859
BINOMIAL DISTRIBUTION
• Mean (Expected Value): The average number of successes we
expect to see over many trials. 𝛍 = 𝐧𝐩
• Variance: The measure of how spread out the possible outcomes
are from the mean. 𝝈𝟐 = 𝐧𝐩𝐪
• Standard Deviation: The square root of the variance, giving us a
measure of spread in the exact same units as our data. 𝛔 = 𝒏𝒑𝒒
BINOMIAL DISTRIBUTION
4. Find the mean and standard deviation of a random variable following a
binomial distribution corresponding to 50 trials each with a probability of
success equal to 0.2.
• Solution:
• n = 50 𝐌𝐞𝐚𝐧 𝐨𝐫 𝐄𝐱𝐩𝐞𝐜𝐭𝐞𝐝 𝐯𝐚𝐥𝐮𝐞: 𝐒𝐭𝐚𝐧𝐝𝐚𝐫𝐝 𝐝𝐞𝐯𝐢𝐚𝐭𝐢𝐨𝐧, 𝛔
μ = np σ = npq
• p = 0.20 μ = 50 0.20 σ = 10 ∗ 0.20 ∗ 0.80
• q = 0.80 𝛍 = 𝟏𝟎 𝛔 = 1.2649
BINOMIAL DISTRIBUTION
5. (from previous problem 1) n = 10 (blocks tested) p = 0.05 (probability of a
block failing) q = 0.95 (probability of a block passing)
• 𝐌𝐞𝐚𝐧 𝐨𝐫 𝐄𝐱𝐩𝐞𝐜𝐭𝐞𝐝 𝐯𝐚𝐥𝐮𝐞:
•μ = np
•μ = 10 0.05
• 𝛍 = 𝟎. 𝟓
• 𝐒𝐭𝐚𝐧𝐝𝐚𝐫𝐝 𝐝𝐞𝐯𝐢𝐚𝐭𝐢𝐨𝐧, 𝛔
•σ = npq
•σ = 10 ∗ 0.05 ∗ 0.95
• 𝛔 = 0.6892
MULTINOMIAL DISTRIBUTION
An extension of the Binomial Distribution. While the Binomial Distribution is
used when there are exactly two possible outcomes (like Pass/Fail), the
Multinomial Distribution is used when a single trial can result in more than
two mutually exclusive outcomes.
RULES
•Multiple Outcomes: Every trial must result in exactly one of several (k)
mutually exclusive outcomes.
•Independent Trials: The outcome of one trial cannot affect the next.
•Fixed Number of Trials: You must know exactly how many total items (n)
you are testing or observing.
•Constant Probabilities: The probabilities of each possible outcome (p1, p2,
p3, etc.) must remain the same for every trial, and all these probabilities
must add up exactly to 1.0 (or 100 percent).
MULTINOMIAL DISTRIBUTION
Formula:
n: The total number of trials.
x1, x2, xk: The specific number of times you want outcome 1, outcome 2, and
outcome k to occur. Note that x1 + x2 + ... + xk must equal n.
p1, p2, pk: The probability of outcome 1, outcome 2, and outcome k
occurring in a single trial.
MULTINOMIAL DISTRIBUTION
6. You are a geotechnical engineer analyzing soil samples from a proposed
highway alignment. Based on preliminary surveys, any single core sample
taken from this site has a 50 percent chance of being predominantly Sand
(p1 = 0.50), a 30 percent chance of being Clay (p2 = 0.30), and a 20 percent
chance of being Silt (p3 = 0.20). You extract 5 random core samples for
testing .What is the probability that your 5 samples consist of exactly 2 Sand,
2 Clay, and 1 Silt?
Solution: 5!
n=5 𝑃 = ∗ 0.5 ∗ 0.3 ∗ 0.2
2! ∗ 2! ∗ 1!
Sand: x1 = 2, p1 = 0.50 𝐏 = 0.135
Clay: x2 = 2, p2 = 0.30
Silt: x3 = 1, p3 = 0.20
Verify rules: x1 + x2 + x3 = 5 (matches n).
p1 + p2 + p3 = 1.0 (valid probabilities).
MULTINOMIAL DISTRIBUTION
• 7. Suppose a bowl has 10 marbles (2 red, 3 green, and 5 blue). 4 marbles
are selected randomly, with replacement. What is the probability of
selecting 2 green marbles and 2 blue marbles?
Solution:
•n = 4
𝐏𝐫𝐨𝐛𝐚𝐛𝐢𝐥𝐢𝐭𝐢𝐞𝐬 𝐟𝐨𝐫 𝐞𝐚𝐜𝐡 𝐱:
• X1 = 2 green
P(green) = 3/10
• X2 = 2 blue P(blue) = 5/10
• X3 = 0 red P(red) = 2/10
𝟒! 𝟐 𝟐 𝟎
𝐏 𝐠𝐫𝒆𝒆𝒏 = 𝟐; 𝒃𝒍𝒖𝒆 = 𝟐 = 𝟎. 𝟑𝟎 𝟎. 𝟓𝟎 𝟎. 𝟐𝟎
𝟐!𝟐!𝟎!
𝐏 𝒈𝒓𝒆𝒆𝒏 = 𝟐; 𝒃𝒍𝒖𝒆 = 𝟐 = 𝟎. 𝟏𝟑𝟓
hypeRgeOMeTRIc DISTRIBUTION
Unlike in binomial distribution where sampling is done WITH replacement
ensuring that individual trials are independent, a HYPERGEOMETRIC
DISTRIBUTION, will use a sample WITHOUTH replacement
Rules:
•The total population size is finite and known (N).
•The population can be split into two distinct groups: "Successes" (K)
and "Failures" (N - K).
•A sample size (n) is chosen without replacement.
•The probability of success changes with each selection because the
trials are dependent
hypeRgeOMeTRIc DISTRIBUTION
The probability distribution for a hypergeometric random variable X, the
number of successes in a random sample size n selected from N items of
which k are labeled success and N-k are labeled as failure, is given as
𝑘𝐶𝑥 (𝑁 − 𝑘)𝐶(𝑛 − 𝑥)
𝑃(𝑋 = 𝑥) =
𝑁𝐶𝑛
N = The total number of items in the population (the whole batch).
K = The total number of successes in the population.
n = The number of items you are choosing for your sample.
x = The specific number of successes you want to find in your sample.
hypeRgeOMeTRIc DISTRIBUTION
8. A crate arrives at a construction site containing 50 high-strength structural
bolts . The quality control team knows from the manufacturer's warning that
exactly 4 of these bolts are defective and do not meet torque specifications.
A site engineer randomly selects 5 bolts from the crate without replacement
to test them. What is the probability that the engineer finds exactly 1
defective bolt in the sample?
Solution: 𝑘𝐶𝑥 (𝑁 − 𝑘)𝐶(𝑛 − 𝑥)
𝑃(𝑋 = 𝑥) =
𝑁𝐶𝑛
N = 50 (total bolts)
K = 4 (defective bolts in the crate) 4𝐶1 46𝐶4
N - K = 46 (good bolts in the crate) 𝑃(𝑋 = 1) =
50𝐶5
n = 5 (bolts selected for the sample)
x = 1 (defective bolts we want to find in our sample) 𝑷 𝑿 = 𝟏 = 𝟎. 𝟑𝟎𝟖𝟏
n - x = 4 (good bolts that must be in our sample)