Year 13 CIE Statistics: Essential Knowledge Review | CIE A Level 统计核心知识点梳理

📚 Year 13 CIE Statistics: Essential Knowledge Review | CIE A Level 统计核心知识点梳理

Statistics at Year 13 under the CIE A Level framework consolidates techniques from S2 while relying on a solid grasp of S1 foundations. This article distils the essential knowledge areas into eleven focused sections, covering probability distributions, continuous random variables, sampling theory, estimation, hypothesis testing, and correlation. Each section is presented with key formulas and concise explanations designed to support revision and deep understanding.

在 CIE A Level 体系中，Year 13 阶段的统计学习以 S2 内容为核心，同时要求对 S1 有牢固的掌握。本文将核心知识点梳理为十一个专题，涵盖概率分布、连续随机变量、抽样理论、估计、假设检验以及相关与回归。每个专题都配有重要公式和简明解说，旨在帮助复习并构建深刻的理解。

1. Probability and Discrete Random Variables | 概率与离散随机变量

A discrete random variable X takes countable values. Its probability distribution satisfies Σ P(X = x) = 1. The expected value is E(X) = Σ x P(X = x), also denoted as the mean μ. The variance is Var(X) = Σ(x − μ)² P(X = x) = E(X²) − [E(X)]². For linear transformations, E(aX + b) = aE(X) + b and Var(aX + b) = a² Var(X).

离散随机变量 X 取可数个值，其概率分布满足 Σ P(X = x) = 1。期望记为 E(X) = Σ x P(X = x)，即均值 μ。方差为 Var(X) = Σ(x − μ)² P(X = x) = E(X²) − [E(X)]²。对于线性变换，E(aX + b) = aE(X) + b，Var(aX + b) = a² Var(X)。

2. The Binomial Distribution | 二项分布

If X ~ B(n, p) with n independent trials and constant probability of success p, then P(X = r) = ⁿCᵣ pʳ qⁿ⁻ʳ where q = 1 − p. The mean is E(X) = np and the variance is Var(X) = npq. The distribution is symmetric when p = 0.5 and may be approximated by a normal distribution if np > 5 and nq > 5.

若 X ~ B(n, p)，即 n 次独立试验，每次成功概率 p 不变，则 P(X = r) = ⁿCᵣ pʳ qⁿ⁻ʳ，其中 q = 1 − p。期望 E(X) = np，方差 Var(X) = npq。当 p = 0.5 时分布对称；当 np > 5 且 nq > 5 时，可用正态分布近似。

3. The Normal Distribution | 正态分布

The normal distribution X ~ N(μ, σ²) is a continuous symmetric distribution. Probabilities are found by standardising: Z = (X − μ) / σ, so Z ~ N(0, 1). The standard normal table gives Φ(z) = P(Z < z). To calculate P(a < X < b), compute Φ((b − μ)/σ) − Φ((a − μ)/σ). In reverse calculations, given a probability p, find the corresponding z-value and convert using X = μ + zσ. Φ(−z) = 1 − Φ(z) is a useful property.

正态分布 X ~ N(μ, σ²) 是连续对称分布。通过标准化 Z = (X − μ) / σ，得到 Z ~ N(0, 1)。标准正态表给出 Φ(z) = P(Z < z)。计算 P(a < X < b) 时，先转换为 Φ((b − μ)/σ) − Φ((a − μ)/σ)。反向计算时，由概率 p 查出 z 值，再由 X = μ + zσ 求得。常用性质 Φ(−z) = 1 − Φ(z)。

4. The Poisson Distribution | 泊松分布

X ~ Po(λ) models the number of events occurring in a fixed interval at a constant average rate λ. Probability is P(X = r) = e⁻λ λʳ / r! for r = 0,1,2,… . The mean and variance are both λ. The Poisson distribution can be used to approximate a binomial B(n, p) when n is large and p is small, taking λ = np. The normal approximation N(λ, λ) is suitable when λ > 15.

X ~ Po(λ) 描述固定区间内以恒定平均速率 λ 发生的事件次数。概率公式 P(X = r) = e⁻λ λʳ / r!（r = 0,1,2,…）。期望和方差均为 λ。当 n 很大且 p 很小时，可用泊松分布近似二项分布 B(n, p)，此时令 λ = np。若 λ > 15，还可以用正态分布 N(λ, λ) 做近似。

5. Continuous Random Variables | 连续随机变量

A continuous random variable X has a probability density function (pdf) f(x) defined on an interval. The total area under f(x) must be 1: ∫_{-∞}^{∞} f(x) dx = 1. The cumulative distribution function (cdf) is F(x) = P(X ≤ x) = ∫_{-∞}^{x} f(t) dt. The median m satisfies F(m) = 0.5; the lower quartile Q₁ satisfies F(Q₁) = 0.25, and the upper quartile Q₃ satisfies F(Q₃) = 0.75. The expected value E(X) = ∫ x f(x) dx, and E(g(X)) = ∫ g(x) f(x) dx. The variance is Var(X) = E(X²) − [E(X)]².

连续随机变量 X 在某个区间上具有概率密度函数 f(x)。f(x) 下的总面积必须为 1：∫_{-∞}^{∞} f(x) dx = 1。累积分布函数 F(x) = P(X ≤ x) = ∫_{-∞}^{x} f(t) dt。中位数 m 满足 F(m) = 0.5；下四分位数 Q₁ 满足 F(Q₁) = 0.25，上四分位数 Q₃ 满足 F(Q₃) = 0.75。期望 E(X) = ∫ x f(x) dx，E(g(X)) = ∫ g(x) f(x) dx。方差 Var(X) = E(X²) − [E(X)]²。

6. Linear Combinations of Random Variables | 随机变量的线性组合

For independent random variables X and Y, the expectation of a linear combination is E(aX + bY) = aE(X) + bE(Y). The variance is Var(aX + bY) = a² Var(X) + b² Var(Y). If the variables are normally distributed, the linear combination is also normally distributed. For example, if X ~ N(μₓ, σₓ²) and Y ~ N(μ_y, σ_y²) are independent, then 3X − 2Y ~ N(3μₓ − 2μ_y, 9σₓ² + 4σ_y²). This result is crucial for working with sample means and sums.

对于相互独立的随机变量 X 和 Y，线性组合的期望为 E(aX + bY) = aE(X) + bE(Y)，方差为 Var(aX + bY) = a² Var(X) + b² Var(Y)。若变量服从正态分布，则其线性组合也服从正态分布。例如 X ~ N(μₓ, σₓ²) 与 Y ~ N(μ_y, σ_y²) 独立时，3X − 2Y ~ N(3μₓ − 2μ_y, 9σₓ² + 4σ_y²)。这一结论对处理样本均值和总和等问题非常关键。

7. Sampling and the Central Limit Theorem | 抽样与中心极限定理

When random samples of size n are taken from a population with mean μ and variance σ², the sample mean X̄ is a random variable with E(X̄) = μ and Var(X̄) = σ² / n. If the population is normally distributed, X̄ also follows a normal distribution. More generally, the Central Limit Theorem (CLT) states that for n ≥ 30, X̄ is approximately normally distributed regardless of the population shape. Thus X̄ ≈ N(μ, σ² / n). The standard error of the mean is σ / √n.

从均值为 μ、方差为 σ² 的总体中抽取容量为 n 的随机样本，样本均值 X̄ 是一个随机变量，满足 E(X̄) = μ，Var(X̄) = σ² / n。若总体服从正态分布，则 X̄ 也服从正态分布。更一般地，中心极限定理指出，当 n ≥ 30 时，无论总体分布形状如何，X̄ 近似服从正态分布，即 X̄ ≈ N(μ, σ² / n)。均值的标准误差为 σ / √n。

8. Estimation and Confidence Intervals | 估计与置信区间

A point estimate for μ is x̄. A 95% confidence interval for the population mean, with known variance σ², is x̄ ± z × σ/√n, where z = 1.96 for 95% confidence. When σ is unknown, use the sample standard deviation s to estimate σ, and the interval becomes x̄ ± t × s/√n with t from the t-distribution and n − 1 degrees of freedom. The width of the interval depends on sample size, variability, and confidence level. For a proportion p from a binomial distribution, a confidence interval is p̂ ± z × √(p̂ q̂ / n) provided np̂ > 5 and n q̂ > 5.

μ 的点估计为样本均值 x̄。在方差 σ² 已知时，总体均值的 95% 置信区间为 x̄ ± z × σ/√n，其中 z = 1.96。当 σ 未知时，用样本标准差 s 估计 σ，区间变为 x̄ ± t × s/√n，t 值来自自由度为 n − 1 的 t 分布。区间宽度取决于样本容量、变异程度和置信水平。对于二项分布的比例 p，在大样本且 np̂ > 5，n q̂ > 5 条件下，置信区间为 p̂ ± z × √(p̂ q̂ / n)。

9. Hypothesis Tests for the Mean (Normal) | 正态均值的假设检验

A hypothesis test for the population mean μ sets H₀: μ = μ₀ against H₁: μ < μ₀, μ > μ₀, or μ ≠ μ₀. When σ is known, the test statistic is Z = (x̄ − μ₀) / (σ/√n). When σ is unknown, use the t-statistic T = (x̄ − μ₀) / (s/√n) with n − 1 degrees of freedom. The critical region is determined by the significance level α. Compare the test statistic with critical value(s) or use the p-value approach: reject H₀ if p < α. A normal approximation may also be used for large-sample tests of a binomial proportion.

对总体均值 μ 的假设检验，原假设 H₀: μ = μ₀，备择假设可选 H₁: μ < μ₀、μ > μ₀ 或 μ ≠ μ₀。当 σ 已知时，检验统计量为 Z = (x̄ − μ₀) / (σ/√n)。σ 未知时，采用 t 统计量 T = (x̄ − μ₀) / (s/√n)，自由度为 n − 1。拒绝域由显著性水平 α 决定。比较检验统计量与临界值，或采用 p 值法：若 p < α，则拒绝 H₀。在大样本时，对二项比例的检验还可使用正态近似。

10. Hypothesis Tests for the Binomial Proportion and Poisson Mean | 二项比例与泊松均值的假设检验

For a binomial parameter p, test H₀: p = p₀ using the exact binomial distribution or a normal approximation Z = (p̂ − p₀) / √(p₀ q₀ / n), provided np₀ > 5 and n q₀ > 5. For a Poisson mean λ, a one-sample test can be performed directly using the Poisson distribution to find the critical region. For large λ (e.g. λ > 15), the normal approximation Z = (x̄ − λ₀) / √(λ₀ / n) for a sample mean, or X ~ Po(λ₀) can be approximated by N(λ₀, λ₀) when an individual count X is used. Always state assumptions and check them.

对于二项参数 p，检验 H₀: p = p₀ 可使用精确二项分布，或在 np₀ > 5 且 n q₀ > 5 时采用正态近似 Z = (p̂ − p₀) / √(p₀ q₀ / n)。对于泊松均值 λ，可直接利用泊松分布确定拒绝域；当 λ 较大（如 λ > 15）时，可使用正态近似。若基于样本均值检验，用 Z = (x̄ − λ₀) / √(λ₀ / n)；若基于单个计数 X，则将 X ~ Po(λ₀) 近似为 N(λ₀, λ₀)。解题时必须陈述假设条件并检验其合理性。

11. Correlation and Linear Regression | 相关与线性回归

Pearson’s product-moment correlation coefficient r measures the strength and direction of a linear relationship between two variables x and y. Its value is given by r = S_xy / √(S_xx S_yy) where S_xx = Σ(x − x̄)², S_yy = Σ(y − ȳ)², S_xy = Σ(x − x̄)(y − ȳ), and −1 ≤ r ≤ 1. The regression line of y on x is y = a + bx, with b = S_xy / S_xx and a = ȳ − b x̄. This line can be used to predict y for a given x, provided the relationship is linear and the data do not extrapolate far outside the observed range.

皮尔逊积矩相关系数 r 衡量变量 x 与 y 之间线性关系的强度和方向，公式为 r = S_xy / √(S_xx S_yy)，其中 S_xx = Σ(x − x̄)²，S_yy = Σ(y − ȳ)²，S_xy = Σ(x − x̄)(y − ȳ)，且 −1 ≤ r ≤ 1。y 对 x 的回归直线为 y = a + bx，斜率 b = S_xy / S_xx，截距 a = ȳ − b x̄。在关系为线性且不超出观测范围太远的前提下，该直线可用于预测给定 x 值时的 y 值。

Published by TutorHao | Statistics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Year 13 CIE Statistics: Essential Knowledge Review | CIE A Level 统计核心知识点梳理