Central Limit Theorem Explained | A-Level进阶数学：中心极限定理详解

📚 Central Limit Theorem Explained | A-Level进阶数学：中心极限定理详解

The Central Limit Theorem (CLT) is one of the most powerful and elegant results in statistics. It explains why the normal distribution appears so frequently in nature and forms the theoretical foundation for many statistical inference procedures. In the Edexcel A-Level Further Mathematics syllabus, a solid grasp of the CLT is essential for tackling problems involving sample means, confidence intervals, and hypothesis testing.

中心极限定理（CLT）是统计学中最为强大且优雅的结论之一。它解释了正态分布为何在自然界中如此频繁出现，也是众多统计推断方法的理论基础。在 Edexcel 进阶数学（A-Level Further Mathematics）大纲中，深刻理解中心极限定理是解决样本均值、置信区间与假设检验问题的关键。

1. What Is the Central Limit Theorem? | 什么是中心极限定理？

The Central Limit Theorem states that when you take a sufficiently large sample of size n from any population with a finite mean μ and finite variance σ², the sampling distribution of the sample mean x̄ will be approximately normal, regardless of the shape of the original population distribution. The approximation improves as the sample size increases.

中心极限定理指出：若从一个具有有限均值 μ 和有限方差 σ² 的总体中抽取容量为 n 的足够大的随机样本，那么样本均值 x̄ 的抽样分布将近似服从正态分布，无论原始总体分布的形状如何。随着样本容量的增大，近似的精度会越来越高。

Mathematically, for a random sample X₁, X₂, …, Xₙ of independent identically distributed variables with mean μ and variance σ², the sample mean x̄ has:

数学上，对于取自同一分布的独立随机样本 X₁, X₂, …, Xₙ（其均值为 μ，方差为 σ²），样本均值 x̄ 具有如下性质：

E(x̄) = μ, Var(x̄) = σ²/n

And for large n, x̄ ~ N(μ, σ²/n) approximately.

并且当 n 较大时，近似有 x̄ ~ N(μ, σ²/n)。

2. Why Is It So Important? | 为什么它如此重要？

The CLT matters for three main reasons. First, it justifies the use of the normal distribution in statistical procedures even when the underlying population is non-normal. Second, it tells us that the mean of a large random sample is concentrated around the population mean, with variability shrinking as n grows. Third, it underpins the construction of confidence intervals and the theory of hypothesis testing.

中心极限定理的重要性体现在三个方面。第一，它说明了即便总体本身不服从正态分布，在统计推断中使用正态分布仍是合理的。第二，它告诉我们大样本均值会集中于总体均值附近，且其变异性随 n 增大而减小。第三，它支撑了置信区间的构建和假设检验理论。

In the Edexcel exam, you are often given a large sample from a population with unknown distribution and asked to use a normal approximation. Without the CLT, these questions would be impossible to solve.

在 Edexcel 考试中，题目常常给出一个来自未知分布总体的大样本，并要求使用正态近似求解。如果没有中心极限定理，这类问题将无从下手。

3. The Exact Distribution of the Sample Mean | 样本均值的精确分布

Before diving into the CLT, it is worth recalling an exact result from the normal distribution. If X₁, X₂, …, Xₙ are independent and each Xᵢ ~ N(μ, σ²), then the sample mean x̄ is exactly normal:

在深入中心极限定理之前，有必要回顾一个关于正态分布的精确结论。若 X₁, X₂, …, Xₙ 相互独立且每个 Xᵢ ~ N(μ, σ²)，那么样本均值 x̄ 精确服从正态分布：

x̄ ~ N(μ, σ²/n)

This exact result holds for any sample size n, because a linear combination of independent normal variables is again normal. It provides a clear benchmark against which the CLT approximation can be compared.

这个精确结论对任何样本容量 n 都成立，因为独立正态变量的线性组合仍然是正态变量。它为我们对比中心极限定理的近似效果提供了一个清晰的标准。

Note that the CLT is a generalisation of this exact result: it relaxes the normality assumption on the parent population but requires a large n to compensate.

注意，中心极限定理是上述精确结果的推广：它放宽了对总体正态性的假设，但要求较大的 n 作为补偿。

4. Conditions and Assumptions | 条件与假设

To apply the CLT safely in exam questions, you must satisfy several conditions:

在考试中安全地应用中心极限定理，需要满足以下若干条件：

Independence: The sample observations must be independent of each other.

独立性：样本观测值之间必须相互独立。
Identical distribution: All observations must come from the same population distribution.

同分布：所有观测值必须来自同一总体分布。
Finite variance: The underlying distribution must have a finite variance σ².

方差有限：总体分布必须具有有限方差 σ²。
Sufficient sample size: The sample size n should be “large enough.” In practice, n ≥ 30 is often cited, but in A-Level further mathematics, the question will tell you when to use the CLT.

样本量充分：样本容量 n 必须”足够大”。实践中常说 n ≥ 30，但在 A-Level 进阶数学中，题目会明确告知何时应使用中心极限定理。

In many exam contexts, the population distribution is discrete (e.g., the number of accidents, or the weight of packages measured to the nearest kilogram). The CLT still applies, but a continuity correction may be necessary if the variable is discrete.

在许多考试情境中，总体分布是离散的（例如事故次数，或精确到千克的包裹重量）。中心极限定理仍然适用，但如果变量是离散的，则可能需要进行连续性校正。

5. Standardising the Sample Mean | 样本均值的标准化

Once we know that x̄ is approximately normal with mean μ and variance σ²/n, we can standardise it to obtain a standard normal Z-score:

已知 x̄ 近似服从均值为 μ、方差为 σ²/n 的正态分布之后，我们可以对其进行标准化，得到标准正态的 Z 分数：

Z = (x̄ − μ) / (σ/√n) ~ N(0, 1)

This transformation is extremely useful, because probabilities of the form P(x̄ < a) can then be found using standard normal probability tables. All Edexcel A-Level exams provide a standard normal distribution table, so you must be comfortable converting between x̄ and Z.

这一变换极为实用，因为形如 P(x̄ < a) 的概率可以通过标准正态分布表求得。Edexcel A-Level 考试均会附赠标准正态分布表，因此你必须熟练掌握在 x̄ 与 Z 之间进行转换。

Consider this typical problem: a random sample of size n = 64 is taken from a population with mean 50 and standard deviation 12. Find the probability that the sample mean exceeds 52. Here, μ = 50, σ = 12, n = 64, so the standard error is 12/√64 = 1.5. Then Z = (52 − 50)/1.5 = 1.333. Using the normal table, P(Z > 1.333) ≈ 0.0912.

看一个典型问题：从均值为 50、标准差为 12 的总体中抽取容量 n = 64 的随机样本，求样本均值超过 52 的概率。这里 μ = 50，σ = 12，n = 64，因此标准误为 12/√64 = 1.5。于是 Z = (52 − 50)/1.5 = 1.333。查正态分布表得 P(Z > 1.333) ≈ 0.0912。

6. The Total of a Random Sample | 随机样本的总和

Sometimes the question asks about the sum of the sample values, rather than the mean. If Xᵢ has mean μ and variance σ², then the total T = X₁ + X₂ + … + Xₙ has mean nμ and variance nσ². For large n, T is approximately normal:

有时问题询问的是样本值之和，而非均值。若 Xᵢ 的均值为 μ、方差为 σ²，则总和 T = X₁ + X₂ + … + Xₙ 的均值为 nμ、方差为 nσ²。对较大的 n，T 近似正态：

T ~ N(nμ, nσ²)

This version of the CLT is crucial for problems involving totals, such as “the total weight of 50 packages” or “the sum of 100 independent measurements.”

这是中心极限定理的一个变体，对于涉及总和的问题至关重要，例如”50 个包裹的总重量”或”100 次独立测量的总和”。

Be careful not to confuse the variance of the mean (σ²/n) with the variance of the total (nσ²). Remembering that means become more precise as n grows, while totals become more variable, can help you avoid this common mistake.

注意不要混淆均值的方差（σ²/n）与总和的方差（nσ²）。理解”均值随 n 增大而更精确，而总和则变异性增大”这一点，有助于避免这个常见错误。

7. Continuity Correction for Discrete Variables | 离散变量的连续性校正

When the underlying variable is discrete but the CLT is used to approximate probabilities via the normal distribution, a continuity correction is essential for accurate results. Suppose X is discrete and we want P(X < 5). The normal approximation would use P(X < 4.5), shifting the boundary by 0.5.

当基础变量是离散的，而又需通过中心极限定理使用正态分布来近似概率时，连续性校正对于获得准确结果至关重要。假设 X 是离散变量，我们想求 P(X < 5)。正态近似应使用 P(X < 4.5)，即将边界平移 0.5。

For the sample mean x̄, if x̄ is based on discrete observations, a continuity correction of 0.5/n should be applied to the boundary of x̄. For example, to approximate P(x̄ < 3) where x̄ is the mean of n = 20 discrete values, compute P(x̄ < 2.975).

对于样本均值 x̄，若 x̄ 基于离散观测值，则应按 0.5/n 对 x̄ 的边界进行连续性校正。例如，要近似 P(x̄ < 3)，其中 x̄ 是 n = 20 个离散值的均值，则应计算 P(x̄ < 2.975)。

In Edexcel A-Level examiners’ reports, candidates frequently lose marks by omitting continuity corrections in discrete-to-normal approximations. Always check whether the variable is discrete before applying the approximation.

在 Edexcel A-Level 阅卷报告中，考生因在离散到正态的近似中遗漏连续性校正而失分的情况屡见不鲜。在应用近似之前，务必确认变量是否为离散型。

8. Worked Example: Mean of an Unknown Distribution | 实例演练：未知分布的样本均值

Let us work through a full exam-style problem step by step.

下面我们逐步完成一道完整的考试风格题目。

Problem: The number of items bought by each customer in a shop has mean 4.5 and standard deviation 3.1. A random sample of 40 customers is selected. Find the probability that the mean number of items bought by the 40 customers is less than 4.

题目：某商店每位顾客购买的商品件数的均值为 4.5，标准差为 3.1。随机抽取 40 名顾客。求这 40 名顾客购买商品件数的均值小于 4 的概率。

Solution: Since n = 40 is large, by the CLT, x̄ is approximately normal with mean μ = 4.5 and standard error σ/√n = 3.1/√40 ≈ 0.4903.

解答：由于 n = 40 较大，由中心极限定理，x̄ 近似服从均值为 μ = 4.5、标准误 σ/√n = 3.1/√40 ≈ 0.4903 的正态分布。

We require P(x̄ < 4). Standardise:

我们要求 P(x̄ < 4)。标准化：

Z = (4 − 4.5) / 0.4903 ≈ −1.019

Therefore P(Z < −1.019) ≈ 0.1541. So the probability is approximately 0.154.

因此 P(Z < −1.019) ≈ 0.1541，即所求概率约为 0.154。

Notice that we never needed to know the exact distribution of the number of items purchased. The CLT made the problem tractable.

注意，我们完全不需要知道购买商品件数的精确分布。中心极限定理使问题得以简化求解。

9. Worked Example: Total of a Sample | 实例演练：样本总和

This example deals with sums of observations.

本例涉及观测值的总和。

Problem: The weight of a randomly selected apple from a farm has mean 150 grams and variance 625 grams². A box contains 36 randomly selected apples. Find the probability that the total weight of the apples exceeds 5.5 kg.

题目：农场随机选取的一个苹果的重量均值为 150 克，方差为 625 克²。一箱装有 36 个随机选取的苹果。求这些苹果总重量超过 5.5 千克的概率。

Solution: Let T be the total weight. Since n = 36 is large, T is approximately normal with:

解答：设 T 为总重量。因 n = 36 较大，T 近似服从正态分布，且：

E(T) = 36 × 150 = 5400 g, Var(T) = 36 × 625 = 22500 g²

Thus SD(T) = √22500 = 150 g. We need P(T > 5500):

因此 SD(T) = √22500 = 150 克。我们需要 P(T > 5500)：

Z = (5500 − 5400) / 150 ≈ 0.667

P(Z > 0.667) ≈ 0.2523. The probability is approximately 0.252.

P(Z > 0.667) ≈ 0.2523，即所求概率约为 0.252。

This example demonstrates the total version of the CLT in action.

本例展示了中心极限定理在总和形式下的应用。

10. Common Misconceptions and Pitfalls | 常见误解与易错点

Several recurring misconceptions appear in A-Level work on the CLT.

在 A-Level 学习中，关于中心极限定理存在几个反复出现的误解。

Misconception: The CLT makes the population distribution normal. It does not. The population distribution stays unchanged; it is the sampling distribution of the mean that becomes approximately normal.

误解一：中心极限定理使总体分布趋于正态。并非如此。总体分布保持不变；趋于正态的是样本均值的抽样分布。
Misconception: The sample size must be exactly 30 or more. There is no universal threshold. The rule of n ≥ 30 is a rough guideline; the speed of convergence depends on the skewness of the underlying distribution.

误解二：样本量必须恰好达到 30 或更多。没有普适的阈值。n ≥ 30 只是一个粗略的准则；收敛速度取决于总体分布的偏斜程度。
Misconception: The mean of the sample equals the population mean exactly. It only equals it in expectation, not in every sample. The sample mean is still a random variable with variability σ²/n.

误解三：样本均值恰好等于总体均值。样本均值仅在期望意义下等于总体均值，而非在每个样本中都相等。样本均值本身仍是一个具有 σ²/n 变异性的随机变量。
Misconception: Continuity correction is optional. For discrete variables, omitting it can change the answer noticeably, and examiners expect it to be applied.

误解四：连续性校正可有可无。对于离散变量，遗漏校正可能使答案产生明显偏差，阅卷者期待考生进行校正。

Being aware of these pitfalls will help you avoid losing easy marks in the examination.

意识到这些易错点，将帮助你在考试中避免不必要的失分。

11. Links with Other Topics | 与其他知识点的联系

The CLT does not exist in isolation within the Edexcel A-Level Further Mathematics specification. It connects naturally to several other key areas.

在 Edexcel A-Level 进阶数学大纲中，中心极限定理并非孤立存在，它与多个关键领域自然相连。

Normal approximation to the binomial distribution: When n is large, a binomial distribution Bin(n, p) can be approximated by a normal distribution with mean np and variance np(1-p). The CLT provides a theoretical justification for this approximation.

二项分布的正态近似：当 n 较大时，二项分布 Bin(n, p) 可用均值为 np、方差为 np(1-p) 的正态分布来近似。中心极限定理为这一近似提供了理论依据。

Normal approximation to the Poisson distribution: For large λ, Po(λ) can be approximated by N(λ, λ). This is again a consequence of the CLT when the Poisson variable is viewed as a sum of independent Poisson increments.

泊松分布的正态近似：当 λ 较大时，Po(λ) 可用 N(λ, λ) 来近似。这同样可以视为中心极限定理的推论，因为泊松变量可看作多个独立泊松增量之和。

Confidence intervals for μ: A 95% confidence interval for the population mean is x̄ ± 1.96σ/√n. This formula rests directly on the CLT. Without the CLT, computing this interval for a non-normal population would be intractable.

总体均值的置信区间：总体均值的 95% 置信区间为 x̄ ± 1.96σ/√n。该公式直接建立在中心极限定理之上。若没有中心极限定理，对非正态总体计算这一区间将极为困难。

Understanding these connections will give you a more cohesive picture of statistics as a whole.

理解这些联系，将使你对统计学的整体图景有更清晰、连贯的认识。

12. Exam Tips and Revision Strategy | 考试技巧与复习策略

Here is a concise checklist to help you approach CLT questions in the exam with confidence.

以下是一份简明清单，帮助你在考试中自信地应对中心极限定理相关题目。

Identify whether the question concerns a mean or a total. Use E(x̄) = μ and Var(x̄) = σ²/n for means; use E(T) = nμ and Var(T) = nσ² for totals.

明确问题是涉及均值还是总和。均值使用 E(x̄) = μ 与 Var(x̄) = σ²/n；总和使用 E(T) = nμ 与 Var(T) = nσ²。
Check whether the variable is discrete. If so, apply a continuity correction of 0.5/n for the mean or 0.5 for the total, as appropriate.

判断变量是否离散。若是，则视情况对均值应用 0.5/n 的连续性校正，或对总和应用 0.5 的校正。
Standardise carefully. Write out Z = (x̄ − μ)/(σ/√n) fully before substituting numbers, to avoid arithmetic errors.

小心标准化。代入数值前完整写出 Z = (x̄ − μ)/(σ/√n)，以避免计算错误。
Use the normal table accurately. Remember whether you want P(Z < z) or P(Z > z), and adjust for symmetry where necessary.

准确查表。注意你需要的是 P(Z < z) 还是 P(Z > z)，必要时利用对称性进行调整。
State the CLT explicitly. In your solution, write “since n is large, by the Central Limit Theorem” to signal your reasoning to the examiner.

明确写出中心极限定理。在解题过程中写出”因为 n 较大，由中心极限定理”，以向阅卷者清楚表明你的推理依据。

When revising, aim to practise at least five past-paper questions on the CLT, including at least one question about a mean and one about a total. This will familiarise you with the standard question formats and common traps.

复习时，至少练习五道关于中心极限定理的历年真题，其中至少一道涉及均值、一道涉及总和。这将使你熟悉标准题型和常见陷阱。

By mastering the Central Limit Theorem, you gain a powerful statistical tool that appears repeatedly throughout the A-Level Further Mathematics curriculum and beyond.

掌握中心极限定理，将使你获得一个强有力的统计工具，它在 A-Level 进阶数学课程及后续学习中会反复出现。

Published by TutorHao | Further Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Central Limit Theorem Explained | A-Level进阶数学：中心极限定理详解