Core Topics in Year 13 Edexcel Statistics | Year 13 Edexcel 统计：核心知识点梳理

📚 Core Topics in Year 13 Edexcel Statistics | Year 13 Edexcel 统计：核心知识点梳理

This article presents a structured overview of the essential topics in Year 13 Edexcel Statistics, covering hypothesis tests for binomial, Poisson and normal distributions, confidence intervals, continuous random variables, linear combinations, chi-squared tests, correlation and regression, probability generating functions and joint distributions. Each section pairs key English explanations with Chinese translations to support bilingual learners.

本文系统梳理了 Year 13 Edexcel 统计学的核心知识点，涵盖二项、泊松和正态分布的假设检验、置信区间、连续随机变量、线性组合、卡方检验、相关与回归、概率母函数以及联合分布。每个要点均以中英文对照呈现，帮助双语学习者巩固理解。

---

1. Hypothesis Testing for a Binomial Proportion | 二项比例假设检验

A hypothesis test for a binomial proportion is used to assess a claim about the population probability p based on an observation from X ∼ B(n, p) . The null hypothesis H₀: p = p₀ is tested against a one‑tailed or two‑tailed alternative. The test statistic is the number of successes, and the critical region is found by summing binomial probabilities until the significance level is reached.

对二项比例的假设检验利用来自 X ∼ B(n, p) 的观测值，评估关于总体概率 p 的断言。原假设 H₀: p = p₀ 针对单尾或双尾备择假设进行检验。检验统计量为成功次数，通过累计二项概率确定临界域，使其不超过显著性水平。

If the observed value falls in the critical region, H₀ is rejected. For a two‑tailed test, the significance level is split equally between the two tails. The actual probability of a Type I error is reported as the actual significance level when using discrete distributions.

若观测值落入临界域，则拒绝 H₀ 。对于双尾检验，显著性水平被等分到两侧。由于离散分布的特性，通常还报告第一类错误的实际概率，即实际显著性水平。

P(X ≤ c) ≤ α/2 or P(X ≥ d) ≤ α/2

---

2. Hypothesis Testing for a Poisson Mean | 泊松均值假设检验

When the data are modelled by a Poisson distribution X ∼ Po(λ) , we can test hypotheses about the mean rate λ . The null hypothesis takes the form H₀: λ = λ₀ , and the test uses Poisson probabilities to determine a critical region for the observed count.

当数据可以用泊松分布 X ∼ Po(λ) 建模时，可对平均发生率 λ 进行假设检验。原假设形式为 H₀: λ = λ₀ ，利用泊松概率为观测计数确定临界域。

For a one‑tailed test with H₁: λ > λ₀ , the critical region is X ≥ c , where c is the smallest integer such that P(X ≥ c) ≤ α . For H₁: λ < λ₀ , the region is X ≤ c with the largest c satisfying P(X ≤ c) ≤ α . Normal approximations are not used here; exact Poisson probabilities are required.

对于 H₁: λ > λ₀ 的单尾检验，临界域为 X ≥ c ，其中 c 是满足 P(X ≥ c) ≤ α 的最小整数。对于 H₁: λ < λ₀ ，临界域为 X ≤ c ，取满足 P(X ≤ c) ≤ α 的最大 c 。此处不采用正态近似，必须使用精确泊松概率。

---

3. Hypothesis Testing for the Mean of a Normal Distribution | 正态均值假设检验

When a population is normally distributed with known variance σ² , we test the mean μ using the sample mean x̄ . The test statistic is Z = ^{x̄ - μ₀}⁄_σ/√n , which follows a standard normal distribution under H₀: μ = μ₀ .

当总体服从正态分布且方差 σ² 已知时，使用样本均值 x̄ 检验均值 μ 。检验统计量为 Z = ^{x̄ - μ₀}⁄_σ/√n ，在原假设 H₀: μ = μ₀ 下服从标准正态分布。

The critical values for a test of size α are taken from the standard normal tables. For a two‑tailed test at the 5% level, the critical region is |Z| > 1.96 . The p‑value is the probability of obtaining a result at least as extreme as the observed value, assuming H₀ is true.

检验的临界值取自标准正态分布表。对于 5% 显著性水平的双尾检验，临界域为 |Z| > 1.96 。p 值是在原假设成立的条件下，获得至少与观测值同样极端结果的概率。

---

4. Confidence Intervals | 置信区间

A confidence interval provides a range of plausible values for an unknown population parameter. For the mean μ of a normal distribution with known variance, a 100(1-α)% confidence interval is given by x̄ ± z_α/2 · σ√n .

置信区间给出了未知总体参数可能的取值范围。对于方差已知的正态总体均值 μ ，一个 100(1-α)% 置信区间为 x̄ ± z_α/2 · σ√n 。

For a population proportion p , the approximate confidence interval based on a large sample is p̂ ± z_α/2 √p̂(1-p̂)n , where p̂ is the sample proportion. The interval is valid when n is large enough to satisfy np̂ > 5 and n(1-p̂) > 5 .

对于总体比例 p ，基于大样本的近似置信区间为 p̂ ± z_α/2 √p̂(1-p̂)n ，其中 p̂ 为样本比例。当 n 足够大，满足 np̂ > 5 且 n(1-p̂) > 5 时，该区间有效。

---

5. Continuous Random Variables | 连续随机变量

A continuous random variable is described by its probability density function f(x) and cumulative distribution function F(x) = P(X ≤ x) = ∫_-∞^x f(t)dt . The total area under the pdf equals 1. Probabilities are found by integrating the pdf over an interval: P(a < X < b) = ∫_a^b f(x)dx .

连续随机变量由其概率密度函数 f(x) 和累积分布函数 F(x) = P(X ≤ x) = ∫_-∞^x f(t)dt 描述。密度函数曲线下的总面积为 1。通过积分求区间概率： P(a < X < b) = ∫_a^b f(x)dx 。

The mean (expected value) is μ = E(X) = ∫_{all x} x f(x)dx , and the variance is σ² = Var(X) = E(X²) - [E(X)]² . The median m satisfies F(m) = 0.5 , and percentiles are found from the inverse cdf.

均值（期望值）为 μ = E(X) = ∫ x f(x)dx ，方差为 σ² = Var(X) = E(X²) - [E(X)]² 。中位数 m 满足 F(m) = 0.5 ，百分位数由逆累积分布函数求得。

---

广告 / Sponsored

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

For random variables X and Y and constants a, b , the expectation is linear: E(aX ± bY) = aE(X) ± bE(Y) . The variance formula involves the covariance: Var(aX ± bY) = a² Var(X) + b² Var(Y) ± 2abCov(X,Y) .

对于随机变量 X 、 Y 及常数 a, b ，期望具有线性性质： E(aX ± bY) = aE(X) ± bE(Y) 。方差公式涉及协方差： Var(aX ± bY) = a² Var(X) + b² Var(Y) ± 2abCov(X,Y) 。

If X and Y are independent, Cov(X,Y) = 0 , so the variance simplifies to Var(aX ± bY) = a² Var(X) + b² Var(Y) . These rules extend to sums of more than two variables. For independent normal variables, any linear combination remains normally distributed.

若 X 与 Y 独立，则 Cov(X,Y) = 0 ，方差简化为 Var(aX ± bY) = a² Var(X) + b² Var(Y) 。这些规则可推广至多个变量的组合。对独立正态变量而言，任何线性组合仍服从正态分布。

---

7. Chi-Squared Tests | 卡方检验

The chi‑squared test is used for frequency data. For a goodness‑of‑fit test, the test statistic is χ² = ∑ ^{(O_i - E_i)2}⁄_Ei , where O_i and E_i are the observed and expected frequencies. Under the null hypothesis, it approximately follows a χ² distribution with ν = k - 1 - p degrees of freedom, where p is the number of estimated parameters.

卡方检验适用于频数数据。拟合优度检验的统计量为 χ² = ∑ ^{(O_i - E_i)2}⁄_Ei ，其中 O_i 和 E_i 分别为观测和预期频数。在原假设下，该统计量近似服从自由度为 ν = k - 1 - p 的 χ² 分布， p 为估计的参数个数。

For a test of independence in a two‑way table, expected frequencies are calculated as E_ij = row total × column totalgrand total . The degrees of freedom are (r-1)(c-1) . The test is valid only when all expected frequencies are at least 5. If the calculated χ² exceeds the critical value, the null hypothesis of independence is rejected.

对于列联表独立性的检验，预期频数由 E_ij = 行合计 × 列合计总合计 计算。自由度为 (r-1)(c-1) 。只有在所有预期频数均不小于 5 时检验才有效。若计算出的 χ² 超过临界值，则拒绝独立性原假设。

---

8. Correlation and Regression | 相关与回归

Pearson’s product‑moment correlation coefficient r measures linear association between two variables x and y . It is calculated as r = ^S_xy⁄_{√Sxx Syy} , where S_xy = ∑ (x_i - x̄)(y_i - ȳ) , S_xx = ∑ (x_i - x̄)² , and S_yy = ∑ (y_i - ȳ)² . The value of r lies between –1 and +1.

皮尔逊积矩相关系数 r 衡量两变量 x 和 y 之间的线性关联程度。计算公式为 r = ^S_xy⁄_{√Sxx Syy} ，其中 S_xy = ∑ (x_i - x̄)(y_i - ȳ) ， S_xx = ∑ (x_i - x̄)² ， S_yy = ∑ (y_i - ȳ)² 。 r 的取值范围为 –1 到 +1。

The least‑squares regression line of y on x is y = a + bx , with b = ^S_xy⁄_Sxx and a = ȳ - bx̄ . Spearman’s rank correlation r_s is a non‑parametric alternative, computed as r_s = 1 - ^{6 ∑ d_i2}⁄_n(n²-1) , where d_i is the difference in ranks. It assesses monotonic, rather than strictly linear, relationships.

y 对 x 的最小二乘回归直线为 y = a + bx ，其中 b = ^S_xy⁄_Sxx ， a = ȳ - bx̄ 。斯皮尔曼秩相关系数 r_s 是一种非参数替代指标，计算公式为 r_s = 1 - ^{6 ∑ d_i2}⁄_n(n²-1) ， d_i 为秩次之差。它衡量单调关系，而不限于严格线性关系。

---

9. Probability Generating Functions | 概率母函数

For a non‑negative integer‑valued random variable X , the probability generating function (PGF) is G_X(t) = E(t^X) = ∑_x=0^∞ t^x P(X=x) . It exists for |t| ≤ 1 and uniquely determines the distribution of X .

对于取非负整数值的随机变量 X ，其概率母函数 (PGF) 定义为 G_X(t) = E(t^X) = ∑_x=0^∞ t^x P(X=x) 。该函数在 |t| ≤ 1 时存在，并能唯一确定 X 的分布。

Moments can be found from derivatives at t=1 : E(X) = G'_X(1) and Var(X) = G''_X(1) + G'_X(1) - [G'_X(1)]² . The PGF of the sum of independent random variables is the product of their individual PGFs: G_X+Y(t) = G_X(t) · G_Y(t) . This property is extremely useful for identifying the distribution of a sum.

矩可通过求导获得： E(X) = G'_X(1) ， Var(X) = G''_X(1) + G'_X(1) - [G'_X(1)]² 。独立随机变量之和的 PGF 等于各变量 PGF 的乘积： G_X+Y(t) = G_X(t) · G_Y(t) 。这一性质在识别和的分布时极为有用。

---

10. Joint Distributions of Discrete Random Variables | 离散联合分布

The joint probability distribution of two discrete random variables X and Y is given by the function P(X=x, Y=y) for all values x, y . This is often displayed in a two‑way table. The marginal distributions of X and Y are obtained by summing the joint probabilities over all values of the other variable.

两个离散随机变量 X 和 Y 的联合概率分布由所有取值 x, y 上的 P(X=x, Y=y) 给出，通常以双向表格表示。通过将联合概率按另一变量的所有取值求和，即可得到 X 和 Y 各自的边缘分布。

Independence holds when P(X=x, Y=y) = P(X=x) · P(Y=y) for all x, y . The covariance is Cov(X,Y) = E(XY) - E(X)E(Y) , where E(XY) = ∑ ∑ xyP(X=x, Y=y) . The correlation coefficient ρ = Cov(X,Y)√Var(X)Var(Y) measures the strength of linear dependence.

独立性成立的条件是所有 x, y 满足 P(X=x, Y=y) = P(X=x) · P(Y=y) 。协方差为 Cov(X,Y) = E(XY) - E(X)E(Y) ，其中 E(XY) = ∑ ∑ xyP(X=x, Y=y) 。相关系数 ρ = Cov(X,Y)√Var(X)Var(Y) 衡量线性相关强度。

---

Published by TutorHao | Statistics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）