📚 Year 13 Cambridge Statistics: Winter Intensive Revision Plan | Year 13 Cambridge 统计：寒假强化复习计划
The winter break offers Year 13 Cambridge students a crucial window to transform scattered knowledge into a cohesive, exam-ready skill set. By following a structured intensive revision plan, you can systematically review every topic, from Poisson distributions to chi-squared tests, while honing your problem-solving speed. This article outlines a 7-week strategy that balances concept reinforcement, targeted practice and full mock exams.
寒假为 Year 13 剑桥学子提供了一个关键窗口，将零散的知识转化为严密且应考自如的能力体系。遵循一份结构化的强化复习计划，你可以系统回顾从泊松分布到卡方检验的每一个主题，同时打磨解题速度。本文提供一份为期 7 周的策略，平衡概念巩固、针对性练习与全真模拟考试。
1. Diagnostic Assessment and Goal Setting | 诊断评估与目标设定
Start by attempting a complete S2 past paper under timed conditions. Mark it honestly and categorise your errors into three buckets: conceptual gaps, careless mistakes, or misinterpretations of the question. This diagnosis will inform your weekly priorities. For instance, if you lost marks on continuous random variable integration, week 3 should receive extra focus.
首先，在定时条件下完成一份完整的 S2 历年真题，如实批改并将错误归类为三类：概念漏洞、粗心错误或题意误读。这一诊断将指引你每周的优先级。例如，如果你在连续随机变量的积分上失分，那么第 3 周应予以额外关注。
A useful diagnostic grid records the question number, topic, mark lost, and root cause. Do not simply write "silly mistake" - force yourself to name the exact slip, such as forgetting the continuity correction or reversing the inequality sign. Re-attempt every lost mark three days later without looking at the mark scheme; only then is the gap genuinely closed.
一张实用的诊断表会记录题号、主题、失分与根本原因。不要只写"粗心"，要强迫自己说清具体的失误，比如忘记连续性修正、或把不等号方向写反。三天后在不看评分标准的情况下重做所有失分题，只有这样才能真正补上漏洞。
2. Mapping the S2 Syllabus and Its Weightings | 梳理 S2 大纲与权重分布
List the full S2 syllabus: Poisson distribution, linear combinations of random variables, continuous random variables, sampling and estimation, hypothesis tests (normal and Poisson means), and chi-squared tests. Note the approximate weighting of each topic in the exam to allocate your time effectively.
列出全部 S2 课程主题：泊松分布、随机变量线性组合、连续随机变量、抽样与估计、假设检验（正态均值与泊松均值）以及卡方检验。记录每个主题在考试中的大致权重，以便有效分配时间。
| S2 Topic | Approx. Weight | Core Skills |
|---|---|---|
| Poisson distribution | 15% | PMF, mean/variance, approximations to and from Poisson |
| Linear combinations of random variables | 10% | E(aX + bY), Var(aX + bY), independence condition |
| Continuous random variables | 20% | PDFs, CDFs, percentiles, E(X), Var(X), Normal PDF |
| Sampling and estimation | 20% | Unbiased estimates, CLT, confidence intervals (z-intervals) |
| Hypothesis tests | 20% | Test for a normal mean, test for a Poisson mean, Type I and II errors |
| Chi-squared tests | 15% | Goodness of fit, contingency tables, degrees of freedom |
Write this table onto a single sheet of A3 and pin it above your desk. Each time you complete a past paper, tick the topics you handled confidently and highlight the ones that cost you marks. Over seven weeks the pattern becomes obvious, and your revision stops being a random walk.
把这张表抄在一张 A3 纸上并贴在书桌前。每完成一套真题，就在你处理得自信的主题上打勾，并标出让你失分的主题。七周之后，规律会变得非常明显，你的复习也就不再是漫无目的的漫游。
3. The Seven-Week Framework at a Glance | 七周框架总览
The plan below moves from core topics to full exam rehearsal. Weeks 1 to 3 rebuild the calculus-heavy content while the mind is fresh; weeks 4 to 6 handle sampling, inference and chi-squared; week 7 is entirely devoted to timed practice and error repair.
下面的计划从核心主题推进到全真演练。第 1 至第 3 周利用头脑最清醒的时段重建计算量大的内容；第 4 至第 6 周处理抽样、推断与卡方；第 7 周完全用于限时训练与错题修复。
| Week | Focus | Deliverable |
|---|---|---|
| 1 | Poisson distribution | 40 mixed Poisson questions |
| 2 | Linear combinations | One-page formula card |
| 3 | Continuous random variables | Full integration drill set |
| 4 | Sampling and estimation | Ten confidence intervals solved |
| 5 | Hypothesis tests | A written conclusion bank |
| 6 | Chi-squared tests | Six full chi-squared solutions |
| 7 | Mock papers and technique | Three timed papers, fully corrected |
Keep the daily load realistic: two focused blocks of 90 minutes, one in the morning for new problem types and one in the late afternoon for review and past-paper questions. Consistency over three weeks beats a single heroic all-nighter every time.
保持每日任务量切合实际：两个 90 分钟的专注时段，上午用来处理新题型，傍晚用来复习与做真题。坚持三周的稳定节奏，永远胜过一次通宵苦战。
4. Week 1: Poisson Distribution Foundations | 第 1 周：泊松分布基础
Week 1 rebuilds the Poisson model from first principles. You must be fluent in the probability mass function, the mean-variance identity, and the conditions under which a Poisson model is appropriate: events occur independently, singly, and at a constant average rate in a given interval.
第 1 周从基本原理重建泊松模型。你必须熟练掌握概率质量函数、均值-方差同一性，以及适用泊松模型的条件：事件独立发生、互不重叠、且在给定区间内以恒定平均速率发生。
P(X = r) = e⁻λ × λʳ ÷ r!，r = 0, 1, 2, … E(X) = Var(X) = λ
Two results appear in almost every paper. The first is the addition rule: if X ~ Po(λ) and Y ~ Po(μ) are independent, then X + Y ~ Po(λ + μ). The second is the approximation chain. A binomial Bin(n, p) with large n and small p is well approximated by Po(np), and a Poisson with large λ (conventionally λ > 10) is approximated by N(λ, λ).
几乎每套卷子都会出现两个结论。第一个是可加性：若 X ~ Po(λ) 与 Y ~ Po(μ) 相互独立，则 X + Y ~ Po(λ + μ)。第二个是近似链。当 n 很大、p 很小时，二项分布 Bin(n, p) 可用 Po(np) 近似；当 λ 较大（通常 λ > 10）时，泊松分布可用 N(λ, λ) 近似。
match the length of the interval in the question, and handling conditional probabilities such as P(X = 3 | P(X = 3) ÷ P(X ≥ 2). Build a 40-question drill set this week and time every ten questions.
注意几个经典陷阱：用正态近似泊松时务必做 ±0.5 的连续性修正；记住 λ 必须按题目所给区间长度进行缩放；以及处理条件概率，例如 P(X = 3 | X ≥ 2) = P(X = 3) ÷ P(X ≥ 2)。本周完成 40 题的专项训练，每十题计时一次。
5. Week 2: Linear Combinations of Random Variables | 第 2 周：随机变量的线性组合
This short topic is disproportionately examinable because it is easy to set and easy to get wrong. The rules you must own are the expectation and variance of a linear combination, with the crucial independence condition attached to the variance.
这个主题篇幅不大，却极为高频，因为它容易出题、也容易出错。你必须牢牢掌握线性组合的期望与方差公式，尤其是方差公式附带的独立条件。
E(aX ± bY) = aE(X) ± bE(Y) Var(aX ± bY) = a²Var(X) + b²Var(Y) （X、Y 独立）
Notice that the variance of a difference is a sum, not a difference. Squaring the coefficients is mandatory, and the minus sign disappears because variance is always non-negative. If independence is not stated, the rule above does not apply and the question must be handled differently, usually by working from the joint distribution or the given covariance.
注意：差的方差是"和"而不是"差"。系数必须平方，负号会消失，因为方差永远非负。如果题目没有说明独立，上述公式就不能使用，此时需要通过联合分布或题给协方差来处理。
When normal variables are combined the result stays normal. If X ~ N(μ₁, σ₁²) and Y ~ N(μ₂, σ₂²) are independent, then aX + bY ~ N(aμ₁ + bμ₂, a²σ₁² + b²σ₂²). The standard exam error is to add standard deviations rather than variances; write the variance line first and take the square root at the very end.
正态变量组合后仍服从正态分布。若 X ~ N(μ₁, σ₁²) 与 Y ~ N(μ₂, σ₂²) 独立，则 aX + bY ~ N(aμ₁ + bμ₂, a²σ₁² + b²σ₂²)。最常见的失分点是直接相加标准差而非方差；请先写出方差的那一行，最后一步再开平方。
6. Week 3: Continuous Random Variables and Integration | 第 3 周：连续随机变量与积分
Week 3 is the mathematical heavy lifting of S2. You need fluent integration, careful handling of piecewise definitions, and total clarity about the difference between a probability density function f(x) and a cumulative distribution function F(x).
第 3 周是 S2 中数学强度最高的部分。你需要熟练的积分技巧、谨慎处理分段定义的函数，并彻底厘清概率密度函数 f(x) 与累积分布函数 F(x) 的区别。
∫ f(x) dx = 1 P(a < X < b) = ∫ₐᵇ f(x) dx E(X) = ∫ x f(x) dx Var(X) = E(X²) − [E(X)]²
5 or F(q) = 0.25. Always check that your integrated density equals 1 before moving on; this single check catches most arithmetic slips.
标准题型是：先求常数 k，再求某个概率，然后求 E(X) 与 Var(X)，最后通过解 F(m) = 0.5 或 F(q) = 0.25 求中位数或百分位数。每算完密度函数都先验证其全积分为 1，这一步能拦住绝大多数计算失误。
form. For a normal distribution with mean μ and variance σ², the density is f(x) = 1 ÷ (σ√2π of a distribution is required, remember that for a density the mode is the value of x that maximises f(x), which usually means differentiating f(x) and setting the derivative to zero.
正态密度函数本身值得记成符号形式。若正态分布的均值为 μ、方差为 σ²，则其密度为 f(x) = 1 ÷ (σ√2π) × e^(−(x − μ)² ÷ (2σ²))。若题目要求众数，记住密度的众数是使 f(x) 最大的 x 值，通常做法是对 f(x) 求导并令导数为零。
7. Week 4: Sampling, Unbiased Estimators and the CLT | 第 4 周：抽样、无偏估计量与中心极限定理
The jump from probability to inference is where many students lose momentum. Begin by distinguishing clearly between a population parameter such as μ or σ² and a sample statistic such as X̄ or s². An estimator is unbiased when its expected value equals the parameter it estimates.
从概率转向统计推断，是许多学生丧失节奏的节点。首先要清晰区分总体参数（如 μ、σ²）与样本统计量（如 X̄、s²）。当估计量的期望值等于它所估计的参数时，该估计量是无偏的。
X̄ ~ N(μ, σ² ÷ n) z = (X̄ − μ) ÷ (σ ÷ √n) CI: x̄ ± z × σ ÷ √n
The Central Limit Theorem is the engine behind every confidence interval here: for a sufficiently large sample, the sample mean is approximately normally distributed regardless of the shape of the population. That is why the same z-based machinery can be applied so widely.
中心极限定理是这里一切置信区间背后的引擎：当样本足够大时，无论总体分布形状如何，样本均值都近似服从正态分布。这正是同一套基于 z 的方法能被广泛使用的原因。
For the 95% confidence interval use z = 1.96, and for 99% use z = 2.576. Interpret the result in context: "We are 95% confident that the true mean lies within this interval." Do not say that there is a 95% rval is random. Ten fully written confidence intervals this week will lock in the wording.
95% 置信区间取 z = 1.96，99% 取 z = 2.576。结论要结合语境表述："我们有 95% 的把握认为真实均值落在该区间内。"不要说"μ 落入该区间的概率是 95%"——参数是固定的，随机的是区间。本周完整写出十个置信区间，就能把这种表述固化下来。
8. Week 5: Hypothesis Tests for Normal and Poisson Means | 第 5 周：正态均值与泊松均值的假设检验
A hypothesis test is a written argument with a fixed structure. State H₀ and H₁, choose the significance level, identify the distribution of the test statistic under H₀, find the critical region, compute the observed value, and finish with a conclusion phrased in the context of the question.
假设检验是有固定结构的论述。写清 H₀ 与 H₁，选定显著性水平，确定检验统计量在 H₀ 下的分布，求出临界区域，计算观测值，最后用题目语境写出一句结论。
For a normal mean with known variance, the test statistic is z = (X̄ − μ₀) ÷ (σ ÷ √n). For a Poisson mean, you usually compare the observed count against the Poisson distribution directly, or use the normal approximation N(λ, λ) when λ is large, remembering the continuity correction.
方差已知的正态均值检验，统计量为 z = (X̄ − μ₀) ÷ (σ ÷ √n)。对于泊松均值的检验，通常直接把观测计数与泊松分布比较；当 λ 较大时可改用正态近似 N(λ, λ)，此时别忘了连续性修正。
Precision in the conclusion earns marks. Write "There is sufficient evidence at the 5% level to reject H₀ and conclude that the mean has changed", never "H₀ is true" or "H₀ is accepted". Also prepare short definitions of Type I and Type II errors, and be ready to state the probability of a Type I error equals the significance level. Build a written conclusion bank this week so the phrasing is automatic in the exam.
结论的准确性直接得分。要写"There is sufficient evidence at the 5% level to reject H₀..."，而不要写"H₀ 为真"或"H₀ 被接受"。同时准备第一类错误与第二类错误的简短定义，并牢记第一类错误的概率等于显著性水平。本周建立一份"结论句式库"，让考场上的表达变成条件反射。
9. Week 6: Chi-Squared Tests and Goodness of Fit | 第 6 周：卡方检验与拟合优度
The chi-squared test compares observed frequencies with expected frequencies under a stated null hypothesis. Every calculation follows the same formula, so the marks are lost in the setup rather than the arithmetic.
卡方检验把观测频数与在给定原假设下的期望频数作比较。所有计算都遵循同一公式，因此失分点几乎都在建模设定阶段，而非算术。
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图