📚 Quick Guide to Statistical Terminology for Year 13 CAIE | CAIE Year 13 统计词汇术语速记指南
Welcome, Year 13 statisticians! Mastering the precise vocabulary of Statistics is half the battle in CAIE Further Mathematics. This guide breaks down essential terms across the core topics – from discrete random variables and Poisson processes to hypothesis testing, confidence intervals, and probability generating functions – into clear, student-friendly explanations. Each entry is paired with memorable shortcuts, helping you decode exam questions and express your reasoning with confidence.
Year 13的统计学家们，欢迎！在CAIE进阶数学中，掌握精确的统计词汇是成功的一半。本指南将涵盖离散随机变量、泊松过程、假设检验、置信区间以及概率生成函数等核心主题的关键术语，转化为清晰易懂的解释。每个术语都配有记忆口诀，帮助你快速破解题意，自信地表达推理过程。
1. Discrete Random Variables & Probability Generating Functions | 离散随机变量与概率生成函数
A discrete random variable X takes a finite or countably infinite set of values. Its probability mass function (PMF) gives P(X = x), and the sum of all probabilities equals 1. The expected value E(X) is the mean, while Var(X) = E(X²) – [E(X)]² measures spread. For two independent variables, E(X+Y) = E(X)+E(Y) and Var(X+Y) = Var(X)+Var(Y).
离散随机变量X取有限或可数无限个值。其概率质量函数（PMF）给出P(X = x)，所有概率之和为1。期望值E(X)是均值，而Var(X) = E(X²) – [E(X)]²衡量离散程度。对于两个独立变量，有E(X+Y) = E(X)+E(Y) 及 Var(X+Y) = Var(X)+Var(Y)。
The probability generating function (PGF) is a compact tool defined as GX(t) = E(tX) = Σ P(X = x) tx for |t| ≤ 1. It encodes all probabilities: the coefficient of tk is P(X = k). Importantly, GX(1) = 1, the first derivative at t=1 gives E(X) = G’X(1), and the second derivative helps find Var(X) = G”X(1) + G’X(1) – [G’X(1)]². For independent X and Y, GX+Y(t) = GX(t) × GY(t). Memorise: ‘PGF sums independent variables by multiplying’.
概率生成函数（PGF）是一个简洁的工具，定义为GX(t) = E(tX) = Σ P(X = x) tx，其中|t| ≤ 1。它包含了所有概率信息：tk的系数就是P(X = k)。重要的是，GX(1) = 1，在t=1处的一阶导数给出E(X) = G’X(1)，二阶导数帮助计算方差：Var(X) = G”X(1) + G’X(1) – [G’X(1)]²。对于独立的X和Y，GX+Y(t) = GX(t) × GY(t)。记忆诀窍：’PGF通过相乘来合并独立变量’。
2. The Poisson Distribution & Its Applications | 泊松分布及其应用
A Poisson distribution models the number of events occurring in a fixed interval of time or space, assuming events happen independently at a constant average rate λ (lambda). The PMF is P(X = x) = (e–λ λx) / x! for x = 0, 1, 2, … Both the mean and variance equal λ. This unique property – E(X) = Var(X) = λ – is a hallmark of the Poisson. If two independent Poisson variables with rates λ₁ and λ₂ are summed, the result is also Poisson with rate λ₁ + λ₂.
泊松分布用于模拟固定时间或空间区间内事件发生的次数，假设事件以恒定平均速率λ（lambda）独立发生。其PMF为 P(X = x) = (e–λ λx) / x!，x = 0, 1, 2, … 均值与方差均等于λ。这一独特性质——E(X) = Var(X) = λ——是泊松分布的特征。若两个速率分别为λ₁和λ₂的独立泊松变量相加，结果仍是泊松分布，速率为 λ₁ + λ₂。
In hypothesis testing, you may be asked to use a Poisson model to calculate the probability of a certain observed frequency under a null assumption. A key phrase: ‘assuming the number of accidents per week follows Po(2.4)’. Remember that the Poisson arises naturally as an approximation to the binomial when n is large and p is small, with λ = np. This connection explains its appearance in rare-event scenarios.
在假设检验中，你可能需要用泊松模型计算在零假设下观察到某个频率的概率。关键句型：’假设每周事故数服从Po(2.4)’。请记住，当n很大而p很小时，泊松分布可作为二项分布的自然近似，此时λ = np。这种联系解释了它在稀有事件场景中频繁出现的原因。
3. Continuous Random Variables & The Exponential Distribution | 连续随机变量与指数分布
A continuous random variable X is described by a probability density function (PDF) f(x) ≥ 0, where the area under the curve over an interval gives the probability. The total area under f(x) is 1, i.e. ∫ f(x) dx = 1 over the domain. The cumulative distribution function (CDF) F(x) = P(X ≤ x) = ∫ f(t) dt from the lower bound to x. E(X) = ∫ x f(x) dx and Var(X) = ∫ x² f(x) dx – [E(X)]².
连续随机变量X由概率密度函数（PDF）f(x) ≥ 0描述，曲线下某区间的面积给出概率。整个定义域上f(x)的总面积为1，即 ∫ f(x) dx = 1。累积分布函数（CDF）F(x) = P(X ≤ x) = 从下界到x的∫ f(t) dt。E(X) = ∫ x f(x) dx，Var(X) = ∫ x² f(x) dx – [E(X)]²。
The exponential distribution is a key continuous distribution. If events occur in a Poisson process at rate λ, then the waiting time T until the first event follows an exponential distribution with PDF f(t) = λ e–λt for t ≥ 0. Its mean is 1/λ and variance is 1/λ². The CDF is F(t) = 1 – e–λt. The memoryless property – P(T > s + t | T > s) = P(T > t) – makes it unique. In exam contexts, you’ll often use the CDF to find probabilities like ‘the probability that the next bus arrives within 5 minutes’.
指数分布是重要的连续分布。如果事件以速率λ的泊松过程发生，那么直到第一个事件的等待时间T遵循指数分布，PDF为 f(t) = λ e–λt，t ≥ 0。其均值为1/λ，方差为1/λ²。CDF为F(t) = 1 – e–λt。无记忆性——P(T > s + t | T > s) = P(T > t)——使其独一无二。在考试中，你常用CDF计算’下一辆公交车在5分钟内到达的概率’这类问题。
4. The Normal Distribution & Sampling | 正态分布与抽样
The normal distribution N(μ, σ²) is the most important continuous model in statistics. Its bell-shaped curve is symmetric about the mean μ, with spread controlled by standard deviation σ. To standardise, use Z = (X – μ) / σ, giving Z ~ N(0, 1²). Always sketch a curve and shade the required area when performing probability calculations. Remember the empirical rule: about 68% of data lies within μ ± σ, 95% within μ ± 2σ, and 99.7% within μ ± 3σ.
正态分布N(μ, σ²)是统计学中最重要的连续模型。其钟形曲线关于均值μ对称，离散程度由标准差σ控制。标准化时使用Z = (X – μ) / σ，得到Z ~ N(0, 1²)。进行概率计算时，务必画出曲线并标出所需面积。记住经验法则：约68%的数据落在μ ± σ内，95%落在μ ± 2σ内，99.7%落在μ ± 3σ内。
When drawing a sample of size n from any population with mean μ and variance σ², the sample mean X̄ has a sampling distribution. By the central limit theorem (CLT), for large n (usually n ≥ 30), X̄ is approximately N(μ, σ²/n). This is the foundation for confidence intervals and hypothesis tests about the population mean. The term ‘standard error’ (SE) means the standard deviation of a sample statistic; for the sample mean, SE = σ/√n.
从均值为μ、方差为σ²的总体中抽取容量为n的样本，样本均值X̄具有抽样分布。根据中心极限定理（CLT），当n较大（通常n ≥ 30）时，X̄近似服从N(μ, σ²/n)。这是关于总体均值的置信区间和假设检验的基础。术语’标准误’（SE）指样本统计量的标准差；对于样本均值，SE = σ/√n。
5. Introduction to Hypothesis Testing | 假设检验导论
Hypothesis testing is a decision-making procedure. You start with a null hypothesis H₀ (the status quo, e.g. μ = 50) and an alternative hypothesis H₁ (what you suspect, e.g. μ > 50). A test statistic is calculated from the sample; if it lies in the critical region (rejection region), you reject H₀. The boundary of the critical region is the critical value. The significance level α (commonly 5% or 1%) is the maximum probability of rejecting H₀ when it is actually true – a Type I error.
假设检验是一种决策程序。首先设立零假设H₀（现状，如 μ = 50）和备择假设H₁（你所怀疑的，如 μ > 50）。从样本计算检验统计量；若它落入拒绝域（临界域），则拒绝H₀。拒绝域的边界是临界值。显著性水平α（常为5%或1%）是当H₀为真时错误拒绝它的最大概率——即第一类错误。
A one-tailed test is used when H₁ specifies a direction (e.g. μ > 50, right-tail; μ < 50, left-tail). A two-tailed test is used when H₁ is simply μ ≠ 50, splitting α into two tails. The p-value is the probability of obtaining a test statistic at least as extreme as the observed one, assuming H₀ is true. If p-value < α, reject H₀. Always conclude in context: "There is sufficient evidence at the 5% level to suggest that..."
当备择假设指明方向时（如 μ > 50 右尾；μ < 50 左尾），使用单尾检验。当H₁只是 μ ≠ 50 时，使用双尾检验，将α平分到双尾。p值是在H₀为真的条件下，得到至少与观测值一样极端的检验统计量的概率。若p值 < α，则拒绝H₀。始终需结合语境下结论："在5%的显著性水平下，有充分证据表明……"
6. Confidence Intervals for Means & Proportions | 均值与比例的置信区间
A confidence interval (CI) provides a range of plausible values for an unknown population parameter. For the population mean μ, when σ is known, a 95% CI is: X̄ ± 1.96 × (σ/√n). The number 1.96 is the z-value cutting off 2.5% in each tail of N(0,1). When σ is unknown and the sample is from a normal population, the t-distribution is used, but in CAIE Further Statistics, σ is usually assumed known or a large-sample normal approximation is applied.
置信区间（CI）给出了未知总体参数的一个合理取值范围。对于总体均值μ，当σ已知时，95% CI为：X̄ ± 1.96 × (σ/√n)。1.96是N(0,1)双尾各截去2.5%的z值。当σ未知且样本来自正态总体时，需用t分布，但在CAIE进阶统计中，常假定σ已知或使用大样本正态近似。
For a population proportion p, based on a sample proportion p̂, the approximate 95% CI is: p̂ ± 1.96 × √[p̂(1 – p̂)/n]. The margin of error captures the width of the interval. Interpreting a 95% CI: if we repeated the sampling many times, about 95% of the intervals would contain the true parameter. Never say ‘there is a 95% probability that the true mean lies in this interval’ – that implies the parameter is random, which it is not.
对于总体比例p，基于样本比例p̂，近似95% CI为：p̂ ± 1.96 × √[p̂(1 – p̂)/n]。误差幅度决定了区间的宽度。解释95% CI的正确说法：如果重复抽样多次，大约95%的区间会包含真实参数。切勿说’真实均值有95%的概率落在这个区间内’——这暗示参数是随机的，而它并非如此。
7. Normal Approximations | 正态近似
When n is large, the binomial distribution Bin(n, p) can be approximated by a normal distribution N(np, np(1 – p)). The rule of thumb: both np ≥ 5 and n(1 – p) ≥ 5. Because the binomial is discrete and the normal is continuous, a continuity correction must be applied. For example, P(X = 15) becomes P(14.5 < Y < 15.5) where Y ~ N(np, np(1 – p)). Similarly, P(X ≤ 15) ≈ P(Y < 15.5).
当n很大时，二项分布Bin(n, p)可用正态分布N(np, np(1 – p))近似。经验法则：np ≥ 5 且 n(1 – p) ≥ 5。由于二项分布是离散的而正态分布是连续的，必须使用连续性校正。例如，P(X = 15)变为 P(14.5 < Y < 15.5)，其中Y ~ N(np, np(1 – p))。类似地，P(X ≤ 15) ≈ P(Y < 15.5)。
The Poisson distribution Po(λ) can also be approximated by a normal distribution N(λ, λ) when λ is large (typically λ > 15). Again, use a continuity correction: P(X ≤ a) ≈ P(Y < a + 0.5). These approximations allow you to perform hypothesis tests on proportions or count data using the normal distribution when exact binomial or Poisson calculations are too cumbersome.
泊松分布Po(λ)当λ较大时（通常λ > 15），也可用正态分布N(λ, λ)近似。同样需使用连续性校正：P(X ≤ a) ≈ P(Y < a + 0.5)。这些近似使得当精确的二项或泊松计算过于繁琐时，你能利用正态分布对比例或计数数据进行假设检验。
8. Goodness of Fit & Chi-Squared Tests | 拟合优度与卡方检验
The chi-squared (χ²) test is used for categorical data. The goodness-of-fit test checks whether observed frequencies match expected frequencies from a specified distribution. The test statistic is χ² = Σ (Oᵢ – Eᵢ)² / Eᵢ, summed over all categories. Under H₀, this statistic follows a χ² distribution with ν = (number of categories – number of estimated parameters – 1) degrees of freedom. A large χ² value indicates poor fit, leading to rejection of H₀.
卡方（χ²）检验用于分类数据。拟合优度检验用来检查观察频数是否符合特定分布下的期望频数。检验统计量为 χ² = Σ (Oᵢ – Eᵢ)² / Eᵢ，对所有类别求和。在H₀下，该统计量服从自由度为ν =（类别数 – 估计参数个数 – 1）的χ²分布。大的χ²值表示拟合不佳，导致拒绝H₀。
The test for independence in a contingency table examines whether two categorical variables are associated. Expected frequencies are calculated as (row total × column total) / grand total. Degrees of freedom: (rows – 1) × (columns – 1). When using a χ² test, ensure all expected frequencies are at least 5; if not, merge categories or use an alternative test. The p-value comes from the right tail of the χ² distribution.
列联表中的独立性检验用于考察两个分类变量是否关联。期望频数按（行合计 × 列合计）/ 总计 计算。自由度为：(行数 – 1) × (列数 – 1)。使用χ²检验时，要确保所有期望频数至少为5；若不满足，则合并类别或改用其他检验。p值取χ²分布的右侧尾部面积。
9. Linear Regression & Correlation | 线性回归与相关
Linear regression models the relationship between a response variable y and an explanatory variable x with the equation y = a + bx, fitted by the least squares method. The slope b = Sxy / Sxx, and intercept a = ȳ – b x̄, where Sxy = Σ(xᵢ – x̄)(yᵢ – ȳ) and Sxx = Σ(xᵢ – x̄)². The residual for a point is observed y minus predicted y; residual plots help check model assumptions.
线性回归用方程 y = a + bx 模拟响应变量y与解释变量x之间的关系，通过最小二乘法拟合。斜率 b = Sxy / Sxx，截距 a = ȳ – b x̄，其中 Sxy = Σ(xᵢ – x̄)(yᵢ – ȳ)，Sxx = Σ(xᵢ – x̄)²。点的残差为观测y值减去预测y值；残差图有助于检查模型假设。
The product moment correlation coefficient (PMCC), denoted r, measures the strength and direction of a linear relationship. r = Sxy / √(Sxx Syy), ranging from –1 to 1. A value close to 1 indicates strong positive correlation, close to –1 strong negative. However, correlation does not imply causation. Hypothesis tests for ρ (population correlation) use the test statistic t = r√(n – 2) / √(1 – r²) against a t-distribution with n – 2 d.f., though the CAIE syllabus often uses a calculator or tables for critical values.
积矩相关系数（PMCC），记作r，衡量线性关系的强度和方向。r = Sxy / √(Sxx Syy)，取值范围–1到1。接近1表示强正相关，接近–1表示强负相关。然而，相关不意味着因果。对ρ（总体相关系数）的假设检验使用检验统计量 t = r√(n – 2) / √(1 – r²)，与自由度为n – 2的t分布比较，尽管CAIE大纲通常使用计算器或查表获取临界值。
10. Errors & Power of a Test | 检验的错误与功效
Two types of error can occur in hypothesis testing. A Type I error happens when H₀ is true but we reject it; the probability is controlled by the significance level α. A Type II error occurs when H₀ is false but we fail to reject it; its probability is denoted β. The power of a test is 1 – β, the probability of correctly rejecting a false H₀. Power increases with larger sample size, larger effect size, and higher α.
假设检验中可能出现两类错误。第一类错误发生在H₀为真却被拒绝时；其概率由显著性水平α控制。第二类错误发生在H₀为假却未能拒绝时；其概率记为β。检验的功效为1 – β，即正确拒绝一个错误H₀的概率。增大样本量、增大效应量、提高α水平均能提高功效。
Calculating power or β often requires specifying a specific alternative value. For example, if H₀: μ = 50 and we want power against μ = 52, we find the probability of falling in the rejection region under the assumption μ = 52. In exam problems, you might compute the probability of Type II error using the normal distribution. A useful table of contrasts:
计算功效或β通常需指定一个具体的备择值。例如，若H₀: μ = 50，我们想计算对μ = 52的功效，就需在μ = 52的假设下求落入拒绝域的概率。在考试题中，你可能会借用正态分布计算第二类错误的概率。一组有用的对比表：
| Concept | 概念 | Definition | 定义 | Symbol | 符号 |
|---|---|---|
| Type I Error | 第一类错误 | Reject true H₀ | 拒绝真实H₀ | α |
| Type II Error | 第二类错误 | Fail to reject false H₀ | 未拒绝虚假H₀ | β |
| Power | 功效 | Correctly reject H₀ | 正确拒绝H₀ | 1 – β |
11. Connecting the Dots: A Terminology Map | 术语联系速查
To help you internalise how the pieces fit together, keep this mental map handy. The table below links the main concepts with their typical notation, required distribution, and a one-sentence clue for quick recall.
为帮助你内化这些知识点如何相互关联，请随时参考这份思维导图。下表将主要概念与典型记号、所需分布以及一句话记忆线索联系起来，便于快速回顾。
| Topic | 主题 | Key Notation | 关键记号 | Distribution Used | 所用分布 | Memory Cue | 记忆线索 |
|---|---|---|---|
| PGF | 概率生成函数 | G(t), derivs at t=1 | N/A | “Derivative at 1 gives E(X)” |
| Mean CI | 均值置信区间 | X̄ ± z*σ/√n | Normal | “z times standard error” |
| Hypothesis test | 假设检验 | H₀, H₁, p-value | Depends on parameter | “If p is low, H₀ must go” |
| Normal approx | 正态近似 | ±0.5 correction | N(μ, σ²) | “Discrete→continuous needs ±0.5” |
| Chi-squared | 卡方 | χ² = Σ(O-E)²/E | χ² (right tail) | “Big gap between O and E raises χ²” |
12. Final Tips for Exam Success | 考试成功最后锦囊
Learn the vocabulary by writing definitions in your own words. Practice identifying the correct distribution from a scenario: ‘rare events’ point to Poisson, ‘sum of a large sample’ to normal, ‘categorical data’ to chi-squared. Always state hypotheses clearly, define variables, and round final answers sensibly. Use diagrams liberally – normal curves, tables, and labelled calculations help you and the examiner follow your logic. With this terminology guide as your companion, you are well-equipped to tackle any CAIE statistics question with precision and clarity.
用你自己的话写出定义来学习术语。练习从情境中识别正确分布：’稀有事件’暗示泊松，’大样本之和’指向正态，’分类数据’用卡方。务必清晰地陈述假设、定义变量，并合理地舍入最终答案。大胆使用图示——正态曲线、表格和带标注的计算过程有助于你和考官理解你的逻辑。以这份术语指南为伴，你将能精确、清晰地应对任何CAIE统计考题。
Published by TutorHao | Statistics 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