Binomial Distribution: Models the number of successes in n independent Bernoulli trials with success probability p.
PMF: P(X = k) = C(n,k) × p^k × (1-p)^(n-k)
Parameters: n (number of trials), p (probability of success)
- Identify n (trials) and p (success probability)
- Use binomial PMF: P(X = k) = C(n,k) × p^k × (1-p)^(n-k)
- For mean: μ = np
- For variance: σ² = np(1-p)
n = 8 (attempts), p = 0.75 (success probability), k = 6 (desired successes)
P(X = 6) = C(8,6) × (0.75)⁶ × (0.25)²
C(8,6) = 8!/(6!×2!) = (8×7)/(2×1) = 28
(0.75)⁶ = 0.177978515625
(0.25)² = 0.0625
P(X = 6) = 28 × 0.177978515625 × 0.0625 = 0.31146240234375
E[X] = μ = np = 8 × 0.75 = 6
Var(X) = σ² = np(1-p) = 8 × 0.75 × 0.25 = 1.5
The probability of making exactly 6 out of 8 free throws is approximately 0.3115.
Expected successful shots: 6, Variance: 1.5
• Binomial PMF: P(X = k) = C(n,k) × p^k × (1-p)^(n-k)
• Mean: E[X] = np
• Variance: Var(X) = np(1-p)
Geometric Distribution: Models the number of trials needed to get the first success in repeated Bernoulli trials.
PMF: P(X = k) = (1-p)^(k-1) × p
Parameters: p (probability of success on each trial)
p = 0.2 (probability of hitting bullseye)
k = 4 (first success on 4th trial)
P(X = 4) = (1-p)^(k-1) × p
= (1-0.2)^(4-1) × 0.2
= (0.8)³ × 0.2
= 0.512 × 0.2 = 0.1024
E[X] = 1/p = 1/0.2 = 5
Var(X) = (1-p)/p² = (1-0.2)/(0.2)² = 0.8/0.04 = 20
The probability that the first bullseye occurs on the fourth throw is 0.1024.
On average, it takes 5 throws to hit the first bullseye.
• Geometric PMF: P(X = k) = (1-p)^(k-1) × p
• Mean: E[X] = 1/p
• Variance: Var(X) = (1-p)/p²
Poisson Distribution: Models the number of events occurring in a fixed interval of time/space.
PMF: P(X = k) = (λ^k × e^(-λ))/k!
Parameter: λ (average rate of occurrence)
λ = 3 (average customers per 10 minutes)
P(X = 5) = (λ^k × e^(-λ))/k!
= (3^5 × e^(-3))/5!
= (243 × 0.049787068367863944)/120
= 12.098257613390638/120 = 0.10081881344492198
P(X ≤ 2) = P(X = 0) + P(X = 1) + P(X = 2)
P(X = 0) = (3^0 × e^(-3))/0! = 1 × 0.049787068367863944/1 = 0.0498
P(X = 1) = (3^1 × e^(-3))/1! = 3 × 0.049787068367863944/1 = 0.1494
P(X = 2) = (3^2 × e^(-3))/2! = 9 × 0.049787068367863944/2 = 0.2240
P(X ≤ 2) = 0.0498 + 0.1494 + 0.2240 = 0.4232
For Poisson: E[X] = Var(X) = λ = 3
The probability of exactly 5 customers arriving is approximately 0.1008.
The probability of no more than 2 customers arriving is approximately 0.4232.
• Poisson PMF: P(X = k) = (λ^k × e^(-λ))/k!
• Mean & Variance: E[X] = Var(X) = λ
• Addition Rule: P(X ≤ a) = Σ P(X = k) for k = 0 to a
Discrete Probability Distribution: Shows all possible values of a discrete random variable and their associated probabilities.
Probability Mass Function (PMF): Function giving P(X = x) for each possible value x.
Support: Set of all possible values the random variable can take.
Normalization: Sum of all probabilities must equal 1.
• Binomial: P(X=k) = C(n,k)p^k(1-p)^(n-k), E[X]=np, Var(X)=np(1-p)
• Geometric: P(X=k) = (1-p)^(k-1)p, E[X]=1/p, Var(X)=(1-p)/p²
• Poisson: P(X=k) = (λ^k e^(-λ))/k!, E[X]=λ, Var(X)=λ
• Hypergeometric: P(X=k) = [C(K,k)×C(N-K,n-k)]/C(N,n)
Hypergeometric Distribution: Models sampling without replacement from a finite population.
PMF: P(X = k) = [C(K,k) × C(N-K,n-k)] / C(N,n)
Parameters: N (population size), K (successes in population), n (sample size)
N = 25 (total balls), K = 10 (red balls), n = 5 (drawn), k = 3 (desired red)
C(K,k) = C(10,3) = 10!/(3!×7!) = 120
C(N-K,n-k) = C(15,2) = 15!/(2!×13!) = 105
C(N,n) = C(25,5) = 25!/(5!×20!) = 53,130
P(X = 3) = [C(10,3) × C(15,2)] / C(25,5)
= (120 × 105) / 53,130
= 12,600 / 53,130 = 0.2372
E[X] = nK/N = 5 × 10/25 = 2
The probability of drawing exactly 3 red balls is approximately 0.2372.
On average, 2 red balls are expected in a sample of 5.
• Hypergeometric PMF: P(X = k) = [C(K,k) × C(N-K,n-k)] / C(N,n)
• Mean: E[X] = nK/N
• Sampling Without Replacement: Each draw affects subsequent probabilities
Complement Rule: P(at least one) = 1 - P(none)
At Least Problems: Often easier to calculate using complement.
This is a binomial problem: n = 20, p = 0.05, 1-p = 0.95
P(X = 2) = C(20,2) × (0.05)² × (0.95)¹⁸
C(20,2) = 20!/(2!×18!) = 190
(0.05)² = 0.0025
(0.95)¹⁸ ≈ 0.3972
P(X = 2) = 190 × 0.0025 × 0.3972 ≈ 0.1887
P(X ≥ 1) = 1 - P(X = 0)
P(X = 0) = C(20,0) × (0.05)⁰ × (0.95)²⁰
= 1 × 1 × (0.95)²⁰ ≈ 0.3585
P(X ≥ 1) = 1 - 0.3585 = 0.6415
E[X] = np = 20 × 0.05 = 1
The probability of exactly 2 defective bulbs is approximately 0.1887.
The probability of at least 1 defective bulb is approximately 0.6415.
On average, 1 bulb is expected to be defective.
• Complement Rule: P(at least one) = 1 - P(none)
• Binomial Distribution: Fixed trials with constant success probability
• Efficiency: Use complement for "at least" problems
Complete Summary: Discrete Probability Distributions
Discrete Random Variable: Takes on a countable number of distinct values.
Probability Mass Function (PMF): Function P(X = x) that gives probability of each value.
Support: Set of all possible values the variable can take.
Normalization: Σ P(X = x) = 1 for all x in support.
Expected Value (Mean): E[X] = Σ [x × P(X = x)]
Variance: Var(X) = E[X²] - (E[X])² = Σ [(x - μ)² × P(X = x)]
Standard Deviation: σ = √Var(X)
Moment Generating Functions: Advanced tool for deriving moments.
Binomial B(n,p): Successes in n trials
Geometric Geo(p): Trials until first success
Poisson Po(λ): Events in fixed interval
Hypergeometric: Sampling without replacement
Identify Distribution: Look for keywords like "fixed number of trials", "first success", "rate of occurrence".
Use Complements: P(X ≥ k) = 1 - P(X < k) often simplifies calculations.
Check Conditions: Verify independence, constant probability, etc.
Verify Results: Probabilities between 0 and 1, sum to 1.
Discrete Probability Distribution: A probability distribution characterized by a probability mass function that assigns probabilities to individual values of a discrete random variable. Unlike continuous distributions, there are gaps between possible values.
Binomial Distribution: Models the number of successes in a fixed number of independent trials with constant probability of success. Used for scenarios like coin flips, pass/fail tests, yes/no surveys.
Geometric Distribution: Models the number of trials needed to achieve the first success in repeated independent Bernoulli trials. Used for "waiting time" problems like finding first defective item.
Poisson Distribution: Models the number of rare events occurring in a fixed interval of time or space. Used for modeling arrivals, defects, accidents, etc.
Hypergeometric Distribution: Models sampling without replacement from a finite population. Used when the probability changes with each draw, unlike binomial.
- Read Carefully: Identify what quantity is being modeled
- Identify Distribution Type: Based on context and constraints
- Extract Parameters: n, p, λ, etc. from problem statement
- Select Appropriate Formula: PMF, CDF, or properties
- Calculate: Apply formula with care
- Verify: Check if answer makes sense
• Binomial: P(X=k) = C(n,k)p^k(1-p)^(n-k), E[X]=np, Var(X)=np(1-p)
• Geometric: P(X=k) = (1-p)^(k-1)p, E[X]=1/p, Var(X)=(1-p)/p²
• Poisson: P(X=k) = (λ^k e^(-λ))/k!, E[X]=λ, Var(X)=λ
• Hypergeometric: P(X=k) = [C(K,k)×C(N-K,n-k)]/C(N,n), E[X]=nK/N
Questions & Answers
Question: How do I know which discrete distribution to use for a given problem? They all seem similar.
Answer: Great question! Here's how to identify each distribution:
- Binomial: Fixed number of trials (n), constant probability (p), interested in number of successes
- Geometric: Repeated trials until FIRST success, interested in number of trials needed
- Poisson: Rate of occurrence (λ), time/space interval, rare events
- Hypergeometric: Sampling WITHOUT replacement from finite population
Keywords to look for:
- "Out of n" or "in a sample of" → Binomial
- "Until first" or "on trial number" → Geometric
- "Average rate" or "per unit time" → Poisson
- "Without replacement" or "finite population" → Hypergeometric
Always consider whether trials are independent and if probability remains constant!
Question: Why do Poisson and Geometric distributions have such different formulas for expected value? Both model rare events.
Answer: While both involve rare events, they model different aspects:
- Poisson: Models COUNT of events in a fixed interval. E[X] = λ (the rate parameter itself).
- Geometric: Models NUMBER OF TRIALS until first success. E[X] = 1/p.
Intuitive explanation:
If success probability is p = 0.1 (10%), then on average you'd expect to wait 1/0.1 = 10 trials for the first success.
For Poisson, if events occur at rate λ = 0.1 per minute, then on average you'd see 0.1 events per minute.
The difference reflects what we're measuring: count of events vs. number of attempts needed.
Question: Why is the variance of Poisson distribution equal to its mean? Isn't that unusual?
Answer: It's not unusual for Poisson - it's actually a defining characteristic!
Why this happens: The Poisson distribution models events that occur independently at a constant average rate. The randomness comes entirely from the timing of events.
Mathematical insight: If λ is small (rare events), both mean and variance are small. If λ is large, both increase proportionally.
Practical significance: If you observe data and find sample mean ≈ sample variance, Poisson might be a good model. If they're very different, consider alternatives.
This relationship makes Poisson both elegant mathematically and useful for modeling real-world phenomena where variation scales with the mean.