3.4.4 Normal Approximation to the Binomial Distribution
3.4.4 Normal Approximation to the Binomial Distribution¶
Key Concepts
- Binomial Distribution: Discrete distribution useful for modeling the number of successes in a fixed number of independent Bernoulli trials.
- Normal Distribution: Continuous distribution often used to approximate the binomial distribution under certain conditions.
Detailed Explanation
Conditions for Normal Approximation
- The sample size n should be large enough such that both np and n(1-p) are greater than or equal to 5.
- This condition ensures that the distribution of the sample proportion is sufficiently symmetrical.
Matching Parameters
- The mean (μ) and variance (σ^2) of the normal distribution are set to np and np(1-p) respectively, to match those of the binomial distribution.
Continuity Correction
- When using the normal approximation, a continuity correction is applied to adjust for the approximation of a discrete distribution by a continuous one. This involves adjusting the binomial variable x by 0.5 in calculations.
Practical Application
Example Calculation:
- Suppose you flip a fair coin 100 times. What is the probability of getting exactly 50 heads?
- Using the normal approximation:
- Mean (μ) = 50, Variance (σ^2) = 25 (since p=0.5)
- To find P(X=50), use P(49.5 < X < 50.5) with the normal approximation.
Using Templates for Calculation:
- Binomial probabilities and their normal approximations can be computed using statistical software or Excel templates, which simplify the process of determining these probabilities.