Course contentsShow
← Mathematics
Lesson 2913 of 3,12157. Common DistributionsFree lesson

Normal Approximation to the Binomial

Using the normal distribution to approximate binomial probabilities when n is large.

Normal Approximation to the Binomial

You'll learn: How to use the normal distribution as a shortcut for calculating binomial probabilities when dealing with many trials.

The Core Idea

When you flip a coin 10 times, computing exact binomial probabilities is manageable. But what if you flip it 1000 times and want to find the probability of getting between 480 and 520 heads? The binomial formula becomes tedious—you'd need to sum 41 separate probability terms!

Fortunately, when n is large, the binomial distribution's histogram starts to look remarkably bell-shaped. This is where the normal distribution becomes a powerful approximation tool.

How It Works

For a binomial random variable X with parameters n (trials) and p (success probability):

  1. Compute the binomial mean and variance:

    • μ = np
    • σ² = np(1-p), so σ = √(np(1-p))
  2. Approximate X with a normal distribution: X ≈ Normal(μ, σ²)

  3. Apply the continuity correction: Since the binomial is discrete but the normal is continuous, adjust boundaries by ±0.5. For example:

    • P(X = 5) ≈ P(4.5 < Normal < 5.5)
    • P(X ≤ 5) ≈ P(Normal ≤ 5.5)

When to Use It

The approximation works best when both np ≥ 10 and n(1-p) ≥ 10. These conditions ensure the distribution isn't too skewed toward either extreme.

Analogy: Think of a staircase (binomial) being approximated by a smooth ramp (normal). When there are many small steps (large n), the ramp fits quite well. The continuity correction accounts for the fact that you're replacing discrete steps with a smooth surface.

Key Takeaway: When n is large and np ≥ 10, n(1-p) ≥ 10, approximate Binomial(n,p) with Normal(np, np(1-p)), using continuity correction for accuracy.