Binomial Distribution: Models the number of successes in n independent Bernoulli trials with constant probability p of success.
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 = E[X] = np
- For variance: σ² = Var(X) = np(1-p)
n = 5 (coin flips), p = 0.5 (probability of heads)
P(X = 3) = C(5,3) × (0.5)³ × (0.5)²
C(5,3) = 5!/(3!×2!) = (5×4)/(2×1) = 10
(0.5)³ = 0.125
(0.5)² = 0.25
P(X = 3) = 10 × 0.125 × 0.25 = 0.3125
E[X] = μ = np = 5 × 0.5 = 2.5
Var(X) = σ² = np(1-p) = 5 × 0.5 × 0.5 = 1.25
SD(X) = σ = √1.25 ≈ 1.118
The probability of getting exactly 3 heads in 5 flips is 0.3125.
Expected number of heads: 2.5, Variance: 1.25
• Binomial PMF: P(X = k) = C(n,k) × p^k × (1-p)^(n-k)
• Mean: E[X] = np
• Variance: Var(X) = np(1-p)
At Least Problems: P(X ≥ k) = P(X = k) + P(X = k+1) + ... + P(X = n)
Complement Rule: P(X ≥ k) = 1 - P(X < k) = 1 - P(X ≤ k-1)
n = 8 (attempts), p = 0.75 (success probability)
P(X = 6) = C(8,6) × (0.75)⁶ × (0.25)²
C(8,6) = C(8,2) = (8×7)/(2×1) = 28
(0.75)⁶ ≈ 0.17798
(0.25)² = 0.0625
P(X = 6) = 28 × 0.17798 × 0.0625 ≈ 0.3115
P(X ≥ 6) = P(X = 6) + P(X = 7) + P(X = 8)
P(X = 7) = C(8,7) × (0.75)⁷ × (0.25)¹ = 8 × 0.13348 × 0.25 ≈ 0.2670
P(X = 8) = C(8,8) × (0.75)⁸ × (0.25)⁰ = 1 × 0.10011 × 1 ≈ 0.1001
P(X ≥ 6) = 0.3115 + 0.2670 + 0.1001 = 0.6786
E[X] = np = 8 × 0.75 = 6
Var(X) = np(1-p) = 8 × 0.75 × 0.25 = 1.5
The probability of making exactly 6 free throws is approximately 0.3115.
The probability of making at least 6 free throws is approximately 0.6786.
The expected number of successful shots is 6.
• Binomial PMF: P(X = k) = C(n,k) × p^k × (1-p)^(n-k)
• At Least Problems: P(X ≥ k) = Σ P(X = i) for i = k to n
• Mean: E[X] = np
Accuracy Rate: The probability that a single test result is correct.
Standard Deviation: SD(X) = √Var(X) = √np(1-p), measures spread in original units.
n = 20 (tests), p = 0.95 (accuracy rate)
P(X = 19) = C(20,19) × (0.95)¹⁹ × (0.05)¹
C(20,19) = 20
(0.95)¹⁹ ≈ 0.3585
(0.05)¹ = 0.05
P(X = 19) = 20 × 0.3585 × 0.05 ≈ 0.3585
P(X = 20) = C(20,20) × (0.95)²⁰ × (0.05)⁰
= 1 × (0.95)²⁰ × 1
= (0.95)²⁰ ≈ 0.3585
Var(X) = np(1-p) = 20 × 0.95 × 0.05 = 0.95
SD(X) = √0.95 ≈ 0.9747
E[X] = np = 20 × 0.95 = 19
The probability of exactly 19 accurate results is approximately 0.3585.
The probability of all 20 results being accurate is approximately 0.3585.
The standard deviation of accurate results is approximately 0.9747.
• Binomial PMF: P(X = k) = C(n,k) × p^k × (1-p)^(n-k)
• Variance: Var(X) = np(1-p)
• Standard Deviation: SD(X) = √Var(X)
Binomial Distribution: A discrete probability distribution of the number of successes in a sequence of n independent experiments.
Bernoulli Trial: A single experiment with two possible outcomes (success/failure).
Independence: The outcome of one trial does not affect the outcome of another.
Constant Probability: The probability of success remains the same across all trials.
• PMF: P(X = k) = C(n,k) × p^k × (1-p)^(n-k)
• Mean: E[X] = μ = np
• Variance: Var(X) = σ² = np(1-p)
• Standard Deviation: SD(X) = σ = √np(1-p)
Defect Rate: The probability that a single item is defective.
Fewer Than: P(X < k) = P(X ≤ k-1), includes all values from 0 to k-1.
n = 50 (items), p = 0.03 (defect rate)
P(X = 2) = C(50,2) × (0.03)² × (0.97)⁴⁸
C(50,2) = (50×49)/(2×1) = 1225
(0.03)² = 0.0009
(0.97)⁴⁸ ≈ 0.2333
P(X = 2) = 1225 × 0.0009 × 0.2333 ≈ 0.2555
P(X < 3) = P(X = 0) + P(X = 1) + P(X = 2)
P(X = 0) = C(50,0) × (0.03)⁰ × (0.97)⁵⁰ = 1 × 1 × (0.97)⁵⁰ ≈ 0.2181
P(X = 1) = C(50,1) × (0.03)¹ × (0.97)⁴⁹ = 50 × 0.03 × (0.97)⁴⁹ ≈ 0.3354
P(X = 2) ≈ 0.2555 (calculated above)
P(X < 3) = 0.2181 + 0.3354 + 0.2555 = 0.8090
E[X] = np = 50 × 0.03 = 1.5
Var(X) = np(1-p) = 50 × 0.03 × 0.97 = 1.455
The probability of exactly 2 defective items is approximately 0.2555.
The probability of fewer than 3 defective items is approximately 0.8090.
The expected number of defective items is 1.5.
• Binomial PMF: P(X = k) = C(n,k) × p^k × (1-p)^(n-k)
• Less Than: P(X < k) = P(X ≤ k-1) = Σ P(X = i) for i = 0 to k-1
• Mean: E[X] = np
Between Inclusive: P(a ≤ X ≤ b) = P(X = a) + P(X = a+1) + ... + P(X = b).
Survey Response: Each respondent's answer is an independent Bernoulli trial.
n = 30 (surveyed), p = 0.6 (support probability)
P(X = 18) = C(30,18) × (0.6)¹⁸ × (0.4)¹²
C(30,18) = C(30,12) ≈ 86,493,225
(0.6)¹⁸ ≈ 0.0000001016
(0.4)¹² ≈ 0.00001678
P(X = 18) ≈ 86,493,225 × 0.0000001016 × 0.00001678 ≈ 0.1474
P(15 ≤ X ≤ 20) = P(X = 15) + P(X = 16) + P(X = 17) + P(X = 18) + P(X = 19) + P(X = 20)
Using binomial probabilities:
P(X = 15) ≈ 0.0489
P(X = 16) ≈ 0.0783
P(X = 17) ≈ 0.1106
P(X = 18) ≈ 0.1474
P(X = 19) ≈ 0.1746
P(X = 20) ≈ 0.1797
P(15 ≤ X ≤ 20) ≈ 0.0489 + 0.0783 + 0.1106 + 0.1474 + 0.1746 + 0.1797 = 0.7395
Var(X) = np(1-p) = 30 × 0.6 × 0.4 = 7.2
SD(X) = √7.2 ≈ 2.683
E[X] = np = 30 × 0.6 = 18
The probability of exactly 18 supporters is approximately 0.1474.
The probability of between 15 and 20 supporters (inclusive) is approximately 0.7395.
The standard deviation of the number of supporters is approximately 2.683.
• Binomial PMF: P(X = k) = C(n,k) × p^k × (1-p)^(n-k)
• Range Probability: P(a ≤ X ≤ b) = Σ P(X = k) for k = a to b
• Standard Deviation: SD(X) = √np(1-p)
Complete Summary: Binomial Probability Distributions
Binomial Distribution: Models the number of successes in n independent Bernoulli trials.
Parameters: n (number of trials), p (probability of success).
Support: X can take integer values from 0 to n.
PMF: P(X = k) = C(n,k) × p^k × (1-p)^(n-k).
Mean: E[X] = μ = np.
Variance: Var(X) = σ² = np(1-p).
Standard Deviation: SD(X) = σ = √np(1-p).
Shape: Depends on p; symmetric when p = 0.5.
Fixed Trials: The number of trials n is fixed in advance.
Independent Trials: Outcome of one trial doesn't affect others.
Constant Probability: Probability of success p remains the same.
Binary Outcomes: Each trial has only two possible outcomes.
Identify Parameters: Always determine n and p first.
Use Complements: P(X ≥ k) = 1 - P(X ≤ k-1) for efficiency.
Range Calculations: P(a ≤ X ≤ b) = Σ P(X = k) for k = a to b.
Check Reasonableness: Verify results make sense in context.
Bernoulli Trial: A random experiment with exactly two possible outcomes ("success" and "failure"), with a fixed probability of success p.
Binomial Distribution: The discrete probability distribution of the number of successes in a sequence of n independent Bernoulli trials.
Combination: C(n,k) = n!/[k!(n-k)!] represents the number of ways to choose k successes from n trials.
Success/Failure: Arbitrary labels for the two possible outcomes of each trial.
Independence: The outcome of one trial does not influence the outcome of any other trial.
- Identify Parameters: Determine n (number of trials) and p (probability of success).
- Verify Conditions: Ensure binomial conditions are met.
- Choose Appropriate Formula: Use PMF for exact probabilities, cumulative for ranges.
- Calculate: Apply the binomial formula carefully.
- Interpret Results: Explain in context of the problem.
• PMF: P(X = k) = C(n,k) × p^k × (1-p)^(n-k)
• Mean: E[X] = μ = np
• Variance: Var(X) = σ² = np(1-p)
• Standard Deviation: SD(X) = σ = √np(1-p)
Questions & Answers
Question: Why is the binomial distribution called "binomial"? What does the name mean?
Answer: The name "binomial" comes from the binomial theorem in algebra:
Etymology:
- "Bi-" means two (like bicycle has two wheels)
- "Nominal" relates to terms or names in mathematics
- Binomial = having two terms
Connection to Statistics:
The binomial distribution arises from the binomial theorem: (a + b)ⁿ = Σ C(n,k)aᵏbⁿ⁻ᵏ
In probability terms: (p + q)ⁿ = Σ C(n,k)pᵏqⁿ⁻ᵏ where q = 1-p
This represents the sum of all possible outcomes of n trials with two possible results each.
So the name reflects both the two-outcome nature of each trial and the mathematical connection to binomial expansions.
Question: What does "independence" really mean in the context of binomial distribution? How do I know if trials are independent?
Answer: Independence in binomial distribution means that the outcome of one trial does not affect the probability of success in any other trial.
Examples of Independence:
- Flipping a fair coin multiple times - each flip is independent
- Rolling a die multiple times - each roll is independent
- Surveying different people - one person's answer doesn't affect another's
Examples of Dependence (NOT Binomial):
- Drawing cards without replacement - probability changes with each draw
- Quality testing where defective items affect remaining items
- Events where previous outcomes influence future probabilities
To check independence: Ask yourself, "Does knowing the result of one trial change the probability of success for another trial?" If yes, the trials are not independent.
Question: In Exercise 5, n=30 was relatively large. Does the binomial distribution still work for large n? Are there any limitations?
Answer: Yes, the binomial distribution works for any positive integer value of n, including very large values. However, there are practical considerations:
Computational Challenges:
- Calculating C(n,k) for large n and k becomes computationally intensive
- For very large n, manual calculation becomes impractical
Approximation Solutions:
For large n, the binomial distribution can be approximated by:
- Normal Distribution: When np and n(1-p) are both ≥ 5
- Poisson Distribution: When n is large and p is small, with np moderate
Statistical Validity:
The binomial distribution remains perfectly valid regardless of n size, as long as the four conditions are met. The distribution is fundamental to probability theory and works mathematically for any n.
In practice, computational tools handle large n values efficiently, so the theoretical distribution remains applicable.