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 \( X \sim B(n, p) \). The null hypothesis \( H_0: p = p_0 \) 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 \( X \sim B(n, p) \) 的观测值，评估关于总体概率 \( p \) 的断言。原假设 \( H_0: p = p_0 \) 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, \( H_0 \) 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 \( H_0 \)。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。

\( P(X \le c) \le \alpha/2 \quad \text{or} \quad P(X \ge d) \le \alpha/2 \)

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

When the data are modelled by a Poisson distribution \( X \sim Po(\lambda) \), we can test hypotheses about the mean rate \( \lambda \). The null hypothesis takes the form \( H_0: \lambda = \lambda_0 \), 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 \( X \sim Po(\lambda) \) 建模时，可对平均发生率 \( \lambda \) 进行假设检验。原假设形式为 \( H_0: \lambda = \lambda_0 \)，利用泊松概率为观测计数确定临界域。

For a one‑tailed test with \( H_1: \lambda > \lambda_0 \), the critical region is \( X \ge c \), where \( c \) is the smallest integer such that \( P(X \ge c) \le \alpha \). For \( H_1: \lambda < \lambda_0 \), the region is \( X \le c \) with the largest \( c \) satisfying \( P(X \le c) \le \alpha \). Normal approximations are not used here; exact Poisson probabilities are required.

对于 \( H_1: \lambda > \lambda_0 \) 的单尾检验，临界域为 \( X \ge c \)，其中 \( c \) 是满足 \( P(X \ge c) \le \alpha \) 的最小整数。对于 \( H_1: \lambda < \lambda_0 \)，临界域为 \( X \le c \)，取满足 \( P(X \le c) \le \alpha \) 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 \( \sigma^2 \), we test the mean \( \mu \) using the sample mean \( \bar{x} \). The test statistic is \( Z = \frac{\bar{x} – \mu_0}{\sigma/\sqrt{n}} \), which follows a standard normal distribution under \( H_0: \mu = \mu_0 \).

当总体服从正态分布且方差 \( \sigma^2 \) 已知时，使用样本均值 \( \bar{x} \) 检验均值 \( \mu \)。检验统计量为 \( Z = \frac{\bar{x} – \mu_0}{\sigma/\sqrt{n}} \)，在原假设 \( H_0: \mu = \mu_0 \) 下服从标准正态分布。

The critical values for a test of size \( \alpha \) 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 \( H_0 \) 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 \( \mu \) of a normal distribution with known variance, a \( 100(1-\alpha)\% \) confidence interval is given by \( \bar{x} \pm z_{\alpha/2} \cdot \frac{\sigma}{\sqrt{n}} \).

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 \( \mu \)，一个 \( 100(1-\alpha)\% \) 置信区间为 \( \bar{x} \pm z_{\alpha/2} \cdot \frac{\sigma}{\sqrt{n}} \)。

For a population proportion \( p \), the approximate confidence interval based on a large sample is \( \hat{p} \pm z_{\alpha/2} \sqrt{\frac{\hat{p}(1-\hat{p})}{n}} \), where \( \hat{p} \) is the sample proportion. The interval is valid when \( n \) is large enough to satisfy \( n\hat{p} > 5 \) and \( n(1-\hat{p}) > 5 \).

对于总体比例 \( p \)，基于大样本的近似置信区间为 \( \hat{p} \pm z_{\alpha/2} \sqrt{\frac{\hat{p}(1-\hat{p})}{n}} \)，其中 \( \hat{p} \) 为样本比例。当 \( n \) 足够大，满足 \( n\hat{p} > 5 \) 且 \( n(1-\hat{p}) > 5 \) 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 \( F(x) = P(X \le x) = \int_{-\infty}^{x} f(t)\,dt \). The total area under the pdf equals 1. Probabilities are found by integrating the pdf over an interval: \( P(a < X < b) = \int_{a}^{b} f(x)\,dx \).

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

The mean (expected value) is \( \mu = E(X) = \int_{all\ x} x f(x)\,dx \), and the variance is \( \sigma^2 = Var(X) = E(X^2) – [E(X)]^2 \). The median \( m \) satisfies \( F(m) = 0.5 \), and percentiles are found from the inverse cdf.

均值（期望值）为 \( \mu = E(X) = \int x f(x)\,dx \)，方差为 \( \sigma^2 = Var(X) = E(X^2) – [E(X)]^2 \)。中位数 \( m \) 满足 \( F(m) = 0.5 \)，百分位数由逆累积分布函数求得。

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: \( E(aX \pm bY) = aE(X) \pm bE(Y) \). The variance formula involves the covariance: \( Var(aX \pm bY) = a^2 Var(X) + b^2 Var(Y) \pm 2ab\,Cov(X,Y) \).

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

If \( X \) and \( Y \) are independent, \( Cov(X,Y) = 0 \), so the variance simplifies to \( Var(aX \pm bY) = a^2 Var(X) + b^2 Var(Y) \). 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 \( Var(aX \pm bY) = a^2 Var(X) + b^2 Var(Y) \)。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 \( \chi^2 = \sum \frac{(O_i – E_i)^2}{E_i} \), where \( O_i \) and \( E_i \) are the observed and expected frequencies. Under the null hypothesis, it approximately follows a \( \chi^2 \) distribution with \( \nu = k – 1 – p \) 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 \( \chi^2 = \sum \frac{(O_i – E_i)^2}{E_i} \)，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 \( \nu = k – 1 – p \) 的 \( \chi^2 \) 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 \( E_{ij} = \frac{\text{row total} \times \text{column total}}{\text{grand total}} \). 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 \( \chi^2 \) 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 \( E_{ij} = \frac{\text{行合计} \times \text{列合计}}{\text{总合计}} \) 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 \( \chi^2 \) 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 \( r = \frac{S_{xy}}{\sqrt{S_{xx} S_{yy}}} \), where \( S_{xy} = \sum (x_i – \bar{x})(y_i – \bar{y}) \), \( S_{xx} = \sum (x_i – \bar{x})^2 \), and \( S_{yy} = \sum (y_i – \bar{y})^2 \). 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 \( r = \frac{S_{xy}}{\sqrt{S_{xx} S_{yy}}} \)，其中 \( S_{xy} = \sum (x_i – \bar{x})(y_i – \bar{y}) \)，\( S_{xx} = \sum (x_i – \bar{x})^2 \)，\( S_{yy} = \sum (y_i – \bar{y})^2 \)。\( r \) 的取值范围为 –1 到 +1。

The least‑squares regression line of \( y \) on \( x \) is \( y = a + bx \), with \( b = \frac{S_{xy}}{S_{xx}} \) and \( a = \bar{y} – b\bar{x} \). Spearman’s rank correlation \( r_s \) is a non‑parametric alternative, computed as \( r_s = 1 – \frac{6 \sum d_i^2}{n(n^2-1)} \), 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 \( b = \frac{S_{xy}}{S_{xx}} \)，\( a = \bar{y} – b\bar{x} \)。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 \( r_s = 1 – \frac{6 \sum d_i^2}{n(n^2-1)} \)，\( d_i \) 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 \( G_X(t) = E(t^X) = \sum_{x=0}^{\infty} t^x P(X=x) \). It exists for \( |t| \le 1 \) 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 \( G_X(t) = E(t^X) = \sum_{x=0}^{\infty} t^x P(X=x) \)。该函数在 \( |t| \le 1 \) 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: \( G_{X+Y}(t) = G_X(t) \cdot G_Y(t) \). 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：\( G_{X+Y}(t) = G_X(t) \cdot G_Y(t) \)。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 \( P(X=x, Y=y) = P(X=x) \cdot P(Y=y) \) for all \( x, y \). The covariance is \( Cov(X,Y) = E(XY) – E(X)E(Y) \), where \( E(XY) = \sum \sum xy\,P(X=x, Y=y) \). The correlation coefficient \( \rho = \frac{Cov(X,Y)}{\sqrt{Var(X)Var(Y)}} \) measures the strength of linear dependence.

独立性成立的条件是所有 \( x, y \) 满足 \( P(X=x, Y=y) = P(X=x) \cdot P(Y=y) \)。协方差为 \( Cov(X,Y) = E(XY) – E(X)E(Y) \)，其中 \( E(XY) = \sum \sum xy\,P(X=x, Y=y) \)。相关系数 \( \rho = \frac{Cov(X,Y)}{\sqrt{Var(X)Var(Y)}} \) 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）

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