📚 Central Limit Theorem and Sampling Distribution of the Sample Mean | 中心极限定理与样本均值的抽样分布
The Central Limit Theorem (CLT) is one of the most powerful and beautiful results in statistics. It explains why the normal distribution appears so frequently in nature and why we can make reliable inferences about a population from a single sample. This article explores the sampling distribution of the sample mean, the statement of the CLT, its conditions, and its practical applications in IB Mathematics.
中心极限定理（CLT）是统计学中最强大、最优美的结论之一。它解释了为什么正态分布在自然界中如此频繁地出现，也解释了为什么我们能从单一随机样本中对总体做出可靠的推断。本文将深入探讨样本均值的抽样分布、中心极限定理的表述、适用条件及其在 IB 数学中的实际应用。
1. What Is a Sampling Distribution? | 什么是抽样分布？
Suppose we take a random sample of size n from a population and compute the sample mean x̄. If we repeat this process many times, we obtain many different sample means. The distribution of all possible sample means, for a fixed sample size n, is called the sampling distribution of the sample mean.
假设我们从总体中抽取一个容量为 n 的随机样本，并计算样本均值 x̄。如果我们重复这个过程很多次，就会得到许多不同的样本均值。对于固定的样本容量 n，所有这些可能的样本均值所构成的分布，称为样本均值的抽样分布。
Key idea: the sample mean itself is a random variable. Before we take the sample, x̄ can take many possible values, so it has its own mean, variance, and shape.
关键思想：样本均值本身就是一个随机变量。在我们抽取样本之前，x̄ 可能取许多不同的值，因此它有自己的均值、方差和分布形态。
2. Mean and Variance of the Sample Mean | 样本均值的均值与方差
Let X₁, X₂, …, Xₙ be independent random variables from a population with mean μ and variance σ². The sample mean is defined as:
设 X₁, X₂, …, Xₙ 是来自均值为 μ、方差为 σ² 的总体的独立随机变量。样本均值定义为：
x̄ = (X₁ + X₂ + … + Xₙ) / n
The expected value of x̄ is equal to the population mean:
x̄ 的期望值等于总体均值：
E(x̄) = μ
The variance of x̄ is the population variance divided by n:
x̄ 的方差等于总体方差除以 n：
Var(x̄) = σ² / n
Therefore the standard deviation of x̄, often called the standard error, is σ/√n. Notice that as n increases, the spread of the sample mean decreases.
因此 x̄ 的标准差通常称为标准误，等于 σ/√n。注意，随着 n 增大，样本均值的离散程度会减小。
3. The Classic Statement of the CLT | 中心极限定理的经典表述
If X₁, X₂, …, Xₙ are independent and identically distributed random variables with mean μ and finite variance σ², then for large n, the sample mean x̄ is approximately normally distributed:
如果 X₁, X₂, …, Xₙ 是独立同分布的随机变量，具有均值 μ 和有限方差 σ²，那么当 n 较大时，样本均值 x̄ 近似服从正态分布：
x̄ ≈ N(μ, σ²/n)
Equivalently, the standardized version follows a standard normal distribution:
等价地，标准化后的形式服从标准正态分布：
Z = (x̄ − μ) / (σ/√n) ≈ N(0, 1)
This is remarkable: the result holds regardless of the shape of the original population distribution.
这非常惊人：无论原始总体分布的形态如何，该结论都成立。
4. Why Does It Matter? | 为什么它如此重要？
The CLT justifies the use of the normal distribution in statistical inference. When we do not know the population distribution, we can still use normal probability calculations for the sample mean, provided the sample size is large enough.
中心极限定理为统计推断中使用正态分布提供了依据。当我们不知道总体分布时，只要样本容量足够大，我们仍然可以针对样本均值使用正态概率计算。
In IB Mathematics, this unlocks the ability to construct confidence intervals and perform hypothesis tests about the population mean using z-scores, even when the population is not normally distributed.
在 IB 数学中，这使我们能够构建置信区间，并使用 z 分数对总体均值进行假设检验，即使总体并非正态分布也可以。
5. How Large Must n Be? | n 需要多大？
There is no universal answer, but a common rule of thumb is n ≥ 30. When the population distribution is symmetric and unimodal, even smaller samples may suffice. When the population is heavily skewed or has outliers, larger samples are needed.
这个问题没有统一的答案，但一个常见的经验法则是 n ≥ 30。当总体分布对称且单峰时，较小的样本也可能足够。当总体严重偏斜或存在异常值时，则需要更大的样本。
Key points to remember:
需要记住的要点：
- n ≥ 30 is a rough guideline, not a theorem.
- If the population is already normal, the sampling distribution is exactly normal for any n.
- If the population is nearly normal, the approximation is good even for n = 10.
- n ≥ 30 是粗略的经验法则，而非定理。
- 如果总体本身是正态的，那么对任意 n，样本均值的抽样分布都精确为正态。
- 如果总体近似正态，即使 n = 10，近似效果也很好。
6. Working Example: Uniform Population | 实例：均匀总体
Consider throwing a fair six-sided die. Let X be the outcome. The population distribution is discrete uniform on {1, 2, 3, 4, 5, 6}, with mean μ = 3.5 and variance σ² = 35/12 ≈ 2.917.
考虑掷一枚均匀的六面骰子。设 X 为点数。总体分布是 {1, 2, 3, 4, 5, 6} 上的离散均匀分布，均值为 μ = 3.5，方差为 σ² = 35/12 ≈ 2.917。
If we roll the die 40 times and compute the sample mean, the CLT tells us that:
如果我们掷骰子 40 次并计算样本均值，中心极限定理告诉我们：
x̄ ≈ N(3.5, 2.917/40) = N(3.5, 0.0729)
The probability that the sample mean exceeds 3.8 can be found by standardizing:
样本均值超过 3.8 的概率可以通过标准化来求得：
Z = (3.8 − 3.5) / √(0.0729) = 0.3 / 0.27 ≈ 1.11
Using the normal distribution, P(x̄ > 3.8) = P(Z > 1.11) ≈ 0.1335.
使用正态分布，P(x̄ > 3.8) = P(Z > 1.11) ≈ 0.1335。
7. The CLT for Sums | 中心极限定理的求和形式
The CLT also applies to the sum of the sample values, Sₙ = X₁ + X₂ + … + Xₙ. For large n:
中心极限定理也适用于样本值的总和 Sₙ = X₁ + X₂ + … + Xₙ。对于较大的 n：
Sₙ ≈ N(nμ, nσ²)
This version is useful when a problem asks about the total weight, total time, or total cost of n independent items.
当问题涉及 n 个独立项目的总重量、总时间或总成本时，这一形式非常有用。
Example: A company packs boxes of apples. The weight of each apple has mean 150 g and standard deviation 20 g. A box contains 50 apples. What is the probability the total weight exceeds 7.6 kg?
示例：一家公司包装苹果箱。每个苹果的重量均值为 150 克，标准差为 20 克。一箱装有 50 个苹果。总重量超过 7.6 千克的概率是多少？
Sₙ ≈ N(50 × 150, 50 × 400) = N(7500, 20000)
Z = (7600 − 7500) / √20000 = 100 / 141.42 ≈ 0.71
So P(Sₙ > 7600) = P(Z > 0.71) ≈ 0.2389.
因此 P(Sₙ > 7600) = P(Z > 0.71) ≈ 0.2389。
8. Continuity Correction for Discrete Data | 离散数据的连续性修正
When the underlying population is discrete, the sample mean is also discrete. The normal approximation can be improved by applying a continuity correction. For example, if asking for P(x̄ > a), use a + 0.5/n; for P(x̄ < a), use a − 0.5/n when working with integer-valued data.
当总体是离散的时候，样本均值也是离散的。使用连续性修正可以改进正态近似。例如，在整数数据中，计算 P(x̄ > a) 时使用 a + 0.5/n；计算 P(x̄ < a) 时使用 a − 0.5/n。
In IB exams, however, the continuity correction is often not required unless explicitly stated. Always read the question carefully.
然而，在 IB 考试中，除非题目明确要求，通常不需要进行连续性修正。务必仔细阅读题目。
9. Comparing the CLT and the Normal Population Case | 中心极限定理与正态总体的比较
If the population itself is normally distributed with mean μ and variance σ², then the sample mean is exactly normal:
如果总体本身服从均值为 μ、方差为 σ² 的正态分布，那么样本均值精确服从正态分布：
x̄ ~ N(μ, σ²/n) exactly, for any n
x̄ ~ N(μ, σ²/n) 精确成立，对任意 n 均成立
In contrast, the CLT is an approximation that improves as n increases. The table below compares the two situations:
相比之下，中心极限定理是一种随着 n 增大而更加精确的近似。下表比较了两种情况：
| Condition | Distribution of x̄ | Exact or Approximate? |
| Population is normal | N(μ, σ²/n) | Exact for all n |
| Population is not normal | N(μ, σ²/n) | Approximate for large n |
条件：总体正态 / 总体非正态；x̄ 的分布：N(μ, σ²/n)；精确或近似：对所有 n 精确 / 大 n 近似。
10. Common Exam Mistakes | 常见考试错误
Students often make the following mistakes:
学生常常犯以下错误：
- Confusing σ with σ/√n when standardizing the sample mean.
- Using the CLT when the sample size is very small and the population is highly skewed.
- Forgetting that the variance of x̄ is σ²/n, not σ².
- Assuming the population must be normal for the CLT to apply.
- 在标准化样本均值时混淆 σ 与 σ/√n。
- 当样本量很小且总体严重偏斜时仍然使用中心极限定理。
- 忘记 x̄ 的方差是 σ²/n，而不是 σ²。
- 误以为中心极限定理要求总体必须为正态分布。
11. Step-by-Step Problem-Solving Strategy | 分步解题策略
Follow these steps for CLT problems:
解决中心极限定理问题请遵循以下步骤：
- Identify the population mean μ and variance σ².
- Determine the sample size n.
- State the approximate distribution of the sample mean: N(μ, σ²/n).
- Convert the required probability into a z-score.
- Use the standard normal table or calculator to find the probability.
- 确定总体均值 μ 和总体方差 σ²。
- 确定样本容量 n。
- 写出样本均值的近似分布：N(μ, σ²/n)。
- 将所求概率转化为 z 分数。
- 使用标准正态分布表或计算器求出概率。
Always write the distribution of x̄ explicitly before doing any calculations. This earns method marks in exams and reduces mistakes.
在做任何计算之前，务必明确写出 x̄ 的分布。这可以在考试中获得方法分，并减少错误。
12. Summary and Final Advice | 总结与最终建议
The Central Limit Theorem is the bridge between probability and inference. It states that regardless of the population shape, the sample mean is approximately normal for large samples, with mean μ and standard error σ/√n.
中心极限定理是连接概率论与统计推断的桥梁。它指出，无论总体形态如何，对于大样本，样本均值近似服从正态分布，均值为 μ，标准误为 σ/√n。
In your IB exam, be confident in identifying when to apply the CLT. Practice with problems involving both continuous and discrete populations, and always state your assumptions clearly.
在 IB 考试中，要自信地判断何时应用中心极限定理。多练习涉及连续总体和离散总体的题目，并始终清晰地写出你的假设。
Master the CLT, and you master the foundation of modern statistics.
掌握中心极限定理，你就掌握了现代统计学的基石。
Published by TutorHao | Mathematics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导