Approximating Binomial Distributions
What you'll learn: You'll learn to use the Central Limit Theorem to approximate binomial probabilities with a normal distribution when the number of trials is large.
The Core Idea
When you have a binomial random variable with a large number of trials n, calculating exact probabilities using the binomial formula becomes computationally intensive. Fortunately, the Central Limit Theorem offers a powerful shortcut: you can approximate the binomial distribution with a normal distribution.
Here's why this works: A binomial random variable X ~ Binomial(n, p) represents the sum of n independent Bernoulli trials. By the CLT, when n is large, this sum behaves approximately like a normal distribution.
How to Apply the Approximation
For X ~ Binomial(n, p):
- Check conditions: The approximation works well when both np ≥ 10 and n(1−p) ≥ 10
- Use normal parameters: Approximate X with a normal distribution having:
- Mean: μ = np
- Standard deviation: σ = √(np(1−p))
- Apply continuity correction: Since you're approximating a discrete distribution with a continuous one, adjust boundaries by ±0.5
- Standardize and find probabilities: Convert to Z-scores and use the standard normal table
Example
Suppose X ~ Binomial(100, 0.3). To find P(X ≤ 35):
- Mean: μ = 100(0.3) = 30
- Standard deviation: σ = √(100(0.3)(0.7)) ≈ 4.58
- With continuity correction: P(X ≤ 35) ≈ P(Y ≤ 35.5) where Y ~ Normal(30, 4.58²)
- Standardize: Z = (35.5 − 30)/4.58 ≈ 1.20
- Use the standard normal table to find the probability
Key Takeaway: When n is large, binomial probabilities can be efficiently approximated using a normal distribution with mean np and standard deviation √(np(1−p)), combined with a continuity correction for better accuracy.