Introduction
Binomial distribution is one of the most important discrete distributions in statistics. It models the number of successes in a fixed number of independent trials, each with the same probability of success.
Binomial Experiment
A binomial experiment is a random experiment with the following properties:
- Consists of n independent Bernoulli trials (n is fixed)
- Each trial has only two possible outcomes: Success (S) or Failure (F)
- All trials are independent (result of one trial doesn’t affect others)
- Probability of success p is constant for each trial
- Random variable X is the total number of successes in n trials
In short: Binomial experiment = repetition of independent Bernoulli trials a finite number of times
Binomial Distribution Definition
Consider n independent Bernoulli trials, where:
- p = probability of success in each trial
- q = 1-p = probability of failure in each trial
- X = number of successes in n trials
The random variable X takes values 0, 1, 2, …, n.
Probability Mass Function (PMF)
The number of ways to get x successes in n trials is $\binom{n}{x} = \frac{n!}{x!(n-x)!}$
Therefore:
$$P(X=x) = \binom{n}{x} p^x q^{n-x}, \quad x=0,1,2,…,n$$
where:
- n = number of trials
- p = probability of success per trial
- q = 1-p = probability of failure
- x = number of successes
Notation: $X \sim B(n,p)$ or $X \sim Binomial(n,p)$
Verification
$$\sum_{x=0}^n P(X=x) = \sum_{x=0}^n \binom{n}{x} p^x q^{n-x} = (p+q)^n = 1^n = 1$$ ✓
Key Properties of Binomial Distribution
Mean (Expected Value)
$$E(X) = \mu = np$$
Proof: Since binomial = sum of n Bernoulli trials, E(X) = n·E(Bernoulli) = n·p
Variance
$$\text{Var}(X) = \sigma^2 = npq = np(1-p)$$
Proof: Var(X) = n·Var(Bernoulli) = n·pq
Standard Deviation
$$\sigma = \sqrt{npq} = \sqrt{np(1-p)}$$
Examples of Binomial Random Variables
- Coin tosses: Number of heads in 10 coin flips
- Quality control: Number of defective items in 100 produced
- Medical: Number of patients recovered in 50 treated with drug
- Marketing: Number of customers who click ad in 1000 impressions
- Exam: Number of correct answers in multiple-choice test
Binomial Distribution Examples
Example 1: Coin Toss
Toss a fair coin 8 times. Let X = number of heads.
Given:
- n = 8
- p = 0.5 (fair coin)
- q = 0.5
a) Probability of exactly 3 heads:
$$P(X=3) = \binom{8}{3} (0.5)^3 (0.5)^5$$
$$= \frac{8!}{3!5!} \times 0.125 \times 0.03125$$
$$= 56 \times 0.00390625 = 0.2188$$
b) Probability of at least 6 heads:
$$P(X \geq 6) = P(X=6) + P(X=7) + P(X=8)$$
$$P(X=6) = \binom{8}{6} (0.5)^8 = 28 \times \frac{1}{256} = 0.1094$$
$$P(X=7) = \binom{8}{7} (0.5)^8 = 8 \times \frac{1}{256} = 0.0313$$
$$P(X=8) = \binom{8}{8} (0.5)^8 = 1 \times \frac{1}{256} = 0.0039$$
$$P(X \geq 6) = 0.1094 + 0.0313 + 0.0039 = 0.1446$$
c) Mean and variance:
$$E(X) = np = 8 \times 0.5 = 4$$
$$\text{Var}(X) = npq = 8 \times 0.5 \times 0.5 = 2$$
$$\sigma = \sqrt{2} = 1.414$$
Example 2: Manufacturing Quality Control
A manufacturing process produces items with 95% non-defective rate. From a batch of 20 items, find probabilities:
Given:
- n = 20
- p = 0.95
- q = 0.05
- X = number of non-defective items
a) All 20 items are non-defective:
$$P(X=20) = \binom{20}{20} (0.95)^{20} (0.05)^0$$
$$= 1 \times 0.358 \times 1 = 0.358$$
b) At least 18 items are non-defective:
$$P(X \geq 18) = P(X=18) + P(X=19) + P(X=20)$$
$$P(X=18) = \binom{20}{18} (0.95)^{18} (0.05)^2 = 190 \times 0.3585 \times 0.0025 = 0.1709$$
$$P(X=19) = \binom{20}{19} (0.95)^{19} (0.05)^1 = 20 \times 0.3774 \times 0.05 = 0.3774$$
$$P(X=20) = 0.358$$
$$P(X \geq 18) = 0.1709 + 0.3774 + 0.358 = 0.9063$$
c) Expected number of non-defective items:
$$E(X) = np = 20 \times 0.95 = 19$$
On average, 19 out of 20 items are non-defective.
Example 3: Drug Efficacy
A drug cures patients with probability 0.7. In a trial of 10 patients:
a) Exactly 8 patients are cured:
$$P(X=8) = \binom{10}{8} (0.7)^8 (0.3)^2$$
$$= 45 \times 0.0576 \times 0.09 = 0.2335$$
b) At most 7 patients are cured:
$$P(X \leq 7) = 1 - P(X \geq 8)$$
$$= 1 - [P(X=8) + P(X=9) + P(X=10)]$$
Calculate each:
- P(X=8) = 0.2335
- P(X=9) = 0.1211
- P(X=10) = 0.0282
$$P(X \leq 7) = 1 - 0.3828 = 0.6172$$
c) Mean and variance:
$$E(X) = np = 10 \times 0.7 = 7$$
$$\text{Var}(X) = npq = 10 \times 0.7 \times 0.3 = 2.1$$
$$\sigma = \sqrt{2.1} = 1.449$$
Example 4: Customer Purchases
30% of website visitors make a purchase. From 100 visitors:
a) Probability exactly 25 make purchases:
$$P(X=25) = \binom{100}{25} (0.3)^{25} (0.7)^{75}$$
This is complex to calculate by hand; typically use statistical software or normal approximation.
b) Mean purchases:
$$E(X) = np = 100 \times 0.3 = 30$$
Expected 30 purchases from 100 visitors.
c) Variance:
$$\text{Var}(X) = npq = 100 \times 0.3 \times 0.7 = 21$$
$$\sigma = \sqrt{21} = 4.58$$
Normal Approximation to Binomial
When n is large and p is close to 0.5, use normal approximation:
$$X \approx N(\mu = np, \sigma^2 = npq)$$
This works well when:
- np ≥ 5 and nq ≥ 5 (rule of thumb)
- Use with continuity correction for better accuracy
When to Use Binomial Distribution
✓ Fixed number of independent trials ✓ Each trial has binary outcome ✓ Constant probability across trials ✓ Interest in number of successes ✓ Quality control applications ✓ Biology and genetics experiments
Properties Summary
| Property | Formula |
|---|---|
| PMF | $P(X=x) = \binom{n}{x} p^x(1-p)^{n-x}$ |
| Mean | $\mu = np$ |
| Variance | $\sigma^2 = np(1-p)$ |
| Std Dev | $\sigma = \sqrt{np(1-p)}$ |
| Support | X ∈ {0, 1, 2, …, n} |
| Notation | X ~ B(n,p) |
Related Distributions
Similar Discrete Distributions:
- Bernoulli Distribution - Single trial with success/failure
- Poisson Distribution - Number of events in fixed interval
- Hypergeometric Distribution - Sampling without replacement
- Negative Binomial Distribution - Trials until r successes
Related Concepts:
- Normal Approximation to Binomial - Approximation for large samples
- Poisson Approximation to Binomial - Approximation for rare events
- Expected Value and Variance - Key statistical measures
Applications
- Quality assurance testing
- Medical trial success rates
- Genetics and inheritance
- Website conversion rates
- Election polling
- Reliability testing
References
-
Montgomery, D.C., & Runger, G.C. (2018). Applied Statistics for Engineers and Scientists (6th ed.). John Wiley & Sons. - Binomial distribution applications in quality control, acceptance sampling, and engineering experiments.
-
Walpole, R.E., Myers, S.L., Myers, S.L., & Ye, K. (2012). Probability & Statistics for Engineers & Scientists (9th ed.). Pearson. - Theoretical foundation of binomial distribution and relationships to normal approximation.