Sample Mean and Variance as Estimators
You'll learn why the sample mean and variance are our go-to tools for estimating unknown population parameters.
The Core Idea
When you want to know a population's true mean (μ) or variance (σ²) but can only collect a sample, you need estimators—formulas that use your sample data to approximate these hidden truths. The sample mean (X̄) naturally estimates μ, and the sample variance (s²) estimates σ².
Think of it like tasting soup: you don't eat the entire pot to judge the flavor—you stir well and taste a spoonful. That spoonful's "flavor profile" estimates the whole pot's taste.
Why These Work
Sample Mean as an Estimator
The sample mean X̄ = (X₁ + X₂ + ... + Xₙ)/n is:
- Unbiased: E[X̄] = μ. On average across many samples, it hits the target perfectly.
- Consistent: By the Law of Large Numbers, as n grows, X̄ converges to μ.
- Efficient: Among unbiased estimators, it has the smallest variance (lowest uncertainty).
Sample Variance as an Estimator
The sample variance s² = Σ(Xᵢ - X̄)²/(n-1) estimates σ² and is:
- Unbiased when we divide by (n-1) instead of n—this corrects for underestimating spread when using X̄ (which itself is estimated from the same data).
- Consistent: As n increases, s² approaches σ².
The (n-1) denominator is called Bessel's correction. It accounts for one "degree of freedom" lost when computing X̄ first.
Why They're "Natural"
These estimators mirror their population counterparts: just as μ is the center of a population, X̄ centers the sample. They're computationally simple, mathematically justified, and behave predictably as sample size grows—making them foundational building blocks.
Key Takeaway: The sample mean and variance are natural, unbiased, consistent estimators of population parameters, providing reliable approximations that improve with larger samples.