📚 A-Level Further Maths Example Responses: Paper 4 Unit FS2 | A-Level 进阶数学：FS2 试卷题型解析
The A-Level Further Mathematics Paper 4 (FS2) unit covers advanced statistics topics such as continuous random variables, hypothesis testing, chi-squared tests, Poisson processes and the gamma distribution. Questions often require a confident command of calculus, distribution theory and statistical inference. In this article, we analyse typical exam question types and provide step-by-step example responses that mirror what examiners expect for full marks.
A-Level 进阶数学试卷 4（FS2）涵盖连续随机变量、假设检验、卡方检验、泊松过程与伽马分布等高等统计内容。题目往往需要灵活运用微积分、分布理论和统计推断。本文将分析常见题型，并提供逐步的示例解答，让你的答题表现更贴近评分标准。
1. Overview of FS2 Exam Structure | FS2 试卷结构概览
The FS2 paper lasts 1 hour 30 minutes and carries 75 marks. Questions frequently combine multiple topics – for instance, you might first find the constant in a probability density function, then compute the median, and later use the distribution for a hypothesis test. Marks are weighted heavily towards clear method and correct use of statistical tables.
FS2 考试时长 1 小时 30 分钟，满分 75 分。题目经常融合多个知识点——比如先求概率密度函数中的常数，接着计算中位数，再用于假设检验。得分的关键在于清晰的步骤和正确使用统计表。
| Topic | Typical Marks |
| Continuous random variables (PDF, CDF, moments) | 10–15 |
| Linear combinations of normal variables | 8–12 |
| Hypothesis testing & confidence intervals | 15–20 |
| Chi-squared tests | 8–12 |
| Exponential & Poisson processes | 8–12 |
| Gamma distribution & applications | 6–10 |
| Type I & Type II error, power | 6–10 |
2. Continuous Random Variables: PDF, CDF, Mean & Variance | 连续随机变量：概率密度函数、分布函数、期望与方差
A favourite FS2 starter asks you to determine an unknown constant in a piecewise PDF, then find cumulative distribution function, median and variance. Always start by setting the total area under the PDF equal to 1.
FS2 开篇常考的是分段概率密度函数中常数的确定，再求分布函数、中位数和方差。第一步一定是利用 PDF 的归一化——总面积为 1。
Example: The continuous random variable X has probability density function f(x) = kx(2 − x) for 0 ≤ x ≤ 2, and f(x) = 0 otherwise.
例题：连续随机变量 X 的概率密度函数为 f(x) = kx(2 − x)，0 ≤ x ≤ 2，其他情况为 0。
(a) Show that k = ¾.
(a) 证明 k = ¾。
∫₀² k x(2 − x) dx = 1 → k ∫₀² (2x − x²) dx = k [x² − x³/3]⁰² = k (4 − 8/3) = (4/3)k = 1 ⇒ k = ¾
(b) Find E(X) and Var(X).
(b) 计算 E(X) 和 Var(X)。
E(X) = ∫₀² x · ¾ x(2 − x) dx = ¾ ∫₀² (2x² − x³) dx = ¾ [2x³/3 − x⁴/4]⁰² = ¾ (16/3 − 4) = ¾ × 4/3 = 1
E(X²) = ∫₀² x² · ¾ x(2 − x) dx = ¾ ∫₀² (2x³ − x⁴) dx = ¾ [x⁴/2 − x⁵/5]⁰² = ¾ (8 − 32/5) = ¾ × 8/5 = 6/5
Var(X) = E(X²) − [E(X)]² = 6/5 − 1 = 1/5 (= 0.2)
Always check that your expectation lies within the domain — here E(X)=1 is midway, which makes sense for a symmetric distribution.
务必检查期望是否落在定义域内——这里 E(X)=1 在中间，对称分布吻合。
3. Linear Combinations of Independent Normal Variables | 独立正态变量的线性组合
If X ~ N(μ₁, σ₁²) and Y ~ N(μ₂, σ₂²) are independent, then aX + bY ~ N(aμ₁ + bμ₂, a²σ₁² + b²σ₂²). This result extends to any number of independent normal variables.
若 X ~ N(μ₁, σ₁²) 且 Y ~ N(μ₂, σ₂²) 独立，则 aX + bY ~ N(aμ₁ + bμ₂, a²σ₁² + b²σ₂²)。这个结论可推广到任意多个独立的正态变量。
Example: The weight of a bag of flour is N(1.5 kg, 0.04² kg²) and sugar is N(2.0 kg, 0.06² kg²). Find the probability that the total weight of 4 bags of flour and 3 bags of sugar exceeds 12 kg.
例题：一袋面粉重量服从 N(1.5, 0.04²)，一袋糖重量服从 N(2.0, 0.06²)。求 4 袋面粉加 3 袋糖的总重量超过 12 kg 的概率。
Let T = 4F + 3S. T ~ N(4×1.5 + 3×2.0, 16×0.04² + 9×0.06²) = N(12.0, 16×0.0016 + 9×0.0036) = N(12.0, 0.0256 + 0.0324) = N(12.0, 0.058)
σ_T = √0.058 ≈ 0.2408 kg
P(T > 12) = P(Z > (12 − 12)/0.2408) = P(Z > 0) = 0.5
Notice how symmetrical the problem becomes when the mean matches the threshold. Always standardise correctly using the standard deviation of the linear combination, not the individual standard deviations.
当均值恰好等于阈值时问题呈现出对称性。始终要用线性组合的标准差进行标准化，而不是各变量的标准差。
4. Hypothesis Testing for the Mean of a Normal Distribution | 正态分布均值的假设检验
When the population variance is unknown and the sample is small, FS2 questions often expect a t‑test. Define H₀ and H₁, calculate the test statistic t = (x̄ − μ₀) / (s/√n), and compare with the critical value from the t‑distribution with n−1 degrees of freedom.
当总体方差未知且样本较小时，FS2 题目通常要求使用 t 检验。明确写出 H₀ 与 H₁，计算检验统计量 t = (x̄ − μ₀)/(s/√n)，并与自由度为 n−1 的 t 分布临界值比较。
Example: A sample of 10 batteries has mean life 24.6 hours and standard deviation 1.2 hours. Test at the 5% significance level whether the population mean is less than 25 hours.
例题：10 个电池的样本平均寿命为 24.6 小时，标准差 1.2 小时。在 5% 显著性水平下检验总体均值是否小于 25 小时。
H₀: μ = 25, H₁: μ < 25 (one‑tailed)
t = (24.6 − 25) / (1.2/√10) = −0.4 / 0.3795 = −1.054
Degrees of freedom = 9. Critical value t₉(5%) = −1.833.
Since −1.054 > −1.833, the test statistic is not in the critical region. Do not reject H₀. There is insufficient evidence at the 5% level that the mean life is less than 25 hours.
Always state the conclusion in context and mention the significance level.
务必结合实际情境给出结论，并提及显著性水平。
5. Confidence Intervals for the Mean | 均值的置信区间
A symmetrical confidence interval for μ when σ is unknown uses the t‑distribution: x̄ ± t_(n−1, α/2) × s/√n. Providing the interval along with a hypothesis test shows complete understanding.
方差未知时，μ 的对称置信区间采用 t 分布：x̄ ± t_(n−1, α/2) × s/√n。同时给出置信区间和假设检验能展现完整的理解。
Using the battery data, construct a 95% confidence interval for the population mean.
沿用电池数据，构造总体均值的 95% 置信区间。
n=10, x̄=24.6, s=1.2, t₉(0.025)=2.262
Interval = 24.6 ± 2.262 × (1.2/√10) = 24.6 ± 2.262 × 0.3795 ≈ 24.6 ± 0.858
95% CI = (23.74, 25.46)
Since the interval contains 25, it agrees with the non‑rejection of H₀ at the 5% level. This dual presentation is highly valued by examiners.
区间包含 25，这与 5% 水平下不拒绝 H₀ 的结果一致。这种双向展示很受阅卷者青睐。
6. Chi‑squared Tests for Independence | 独立性卡方检验
Chi‑squared tests in FS2 often involve 2×2 or 3×2 contingency tables. State hypotheses, compute expected frequencies, calculate the test statistic Σ (O − E)²/E, and compare with the critical χ² value at (r−1)(c−1) degrees of freedom.
FS2 的卡方检验常涉及 2×2 或 3×2 列联表。先给出假设，计算期望频数，算检验统计量 Σ (O − E)²/E，并与自由度为 (r−1)(c−1) 的 χ² 临界值比较。
Example: A survey of 200 people checks whether gender and preference for tea or coffee are independent.
| Tea | Coffee | Total | |
| Male | 45 | 55 | 100 |
| Female | 65 | 35 | 100 |
| Total | 110 | 90 | 200 |
Expected: Male‑Tea = 100×110/200 = 55, Male‑Coffee = 45, Female‑Tea = 55, Female‑Coffee = 45.
χ² = (45−55)²/55 + (55−45)²/45 + (65−55)²/55 + (35−45)²/45 = 100/55 + 100/45 + 100/55 + 100/45 ≈ 1.818 + 2.222 + 1.818 + 2.222 = 8.08
df = 1, critical value χ²₁(5%) = 3.841. 8.08 > 3.841, reject H₀. There is evidence of association.
Always check that at least 80% of expected frequencies are above 5 and none are below 1. For 2×2 tables, Yates’ correction is sometimes required depending on the specification.
务必检查至少 80% 的期望频数大于 5 且无一小于 1。部分考纲对 2×2 表格要求 Yates 修正，需确认。
7. Exponential Distribution and Poisson Processes | 指数分布与泊松过程
The exponential distribution models waiting times between independent events. Its PDF is f(x) = λe⁻λˣ for x > 0, mean = 1/λ, and the CDF is F(x) = 1 − e⁻λˣ. When events follow a Poisson process with rate λ per unit time, the number of events in time t is Poisson(λt).
指数分布用于建模独立事件间的等待时间。概率密度函数 f(x) = λe⁻λˣ (x>0)，均值 1/λ，分布函数 F(x) = 1 − e⁻λˣ。当事件遵循单位时间速率 λ 的泊松过程时，时间 t 内的事件数服从 Poisson(λt)。
Example: The lifetime of a component (in hours) is exponential with mean 800. (a) Find the probability it lasts more than 1000 hours. (b) Out of 5 such components, find the probability at least 4 last more than 1000 hours.
例题：某元件寿命（小时）服从均值为 800 的指数分布。(a) 求寿命超过 1000 小时的概率。(b) 若使用 5 个该元件，求至少有 4 个寿命超过 1000 小时的概率。
λ = 1/800. P(X > 1000) = e⁻λ·¹⁰⁰⁰ = e⁻¹·²⁵ = 0.2865 (using tables).
Let Y = number out of 5 that exceed 1000 hours. Y ~ B(5, 0.2865). P(Y ≥ 4) = P(Y=4) + P(Y=5) = 5C4(0.2865)⁴(0.7135) + (0.2865)⁵ ≈ 5×0.00676×0.7135 + 0.00194 ≈ 0.0241 + 0.0019 = 0.0260
The exponential distribution possesses the memoryless property: P(X > s+t | X > s) = P(X > t). This is often tested in conceptual questions.
指数分布具有无记忆性：P(X > s+t | X > s) = P(X > t)。这一性质常在概念题中考查。
8. The Gamma Distribution | 伽马分布
If X₁, X₂, …, Xₙ are independent Exp(λ), then their sum S = ΣXᵢ follows Gamma(n, λ). The general gamma PDF is f(x) = (λᵅ xᵅ⁻¹ e⁻λˣ) / Γ(α) for x > 0, α > 0, λ > 0. Its mean is α/λ, variance is α/λ².
若 X₁, …, Xₙ 相互独立且服从 Exp(λ)，则其和 S = ΣXᵢ 服从 Gamma(n, λ)。一般伽马分布的概率密度函数为 f(x) = (λᵅ xᵅ⁻¹ e⁻λˣ) / Γ(α)，x>0，α, λ>0。均值为 α/λ，方差为 α/λ²。
Example: The total repair time for 3 independent faults, each exponential with mean 30 minutes, follows Gamma(3, 1/30). Find the probability the total repair time exceeds 2 hours (120 minutes).
例题：3 个独立故障的维修时间均服从均值 30 分钟的指数分布，则总维修时间服从 Gamma(3, 1/30)。求总维修时间超过 120 分钟的概率。
α = 3, λ = 1/30. P(S > 120) can be evaluated using the connection to Poisson: P(S > t) = Σₖ₌₀⁴⁻¹ e⁻λt (λt)ᵏ/k! with k=0,1,2.
λt = (1/30)×120 = 4. P(S > 120) = e⁻⁴ (1 + 4 + 4²/2!) = e⁻⁴ × (1+4+8) = 13 e⁻⁴ ≈ 13 × 0.01832 = 0.238
This Poisson‑gamma duality is extremely useful and avoids integrating the gamma PDF directly.
泊松-伽马对偶关系十分实用，避免了直接积分伽马密度函数的繁琐。
9. Type I & Type II Errors and Power | 第一类错误、第二类错误与功效
FS2 candidates must be able to compute the probabilities of Type I error (rejecting H₀ when true) and Type II error (failing to reject H₀ when a specific H₁ is true), as well as power = 1 − P(Type II error).
FS2 考生需要能够计算第一类错误概率（H₀ 为真时拒绝）、第二类错误概率（特定 H₁ 为真时未拒绝）以及功效 = 1 − P(第二类错误)。
Example: Test H₀: μ = 60 against H₁: μ < 60, with known σ = 8, n = 36. The critical region is x̄ < 57.8.
例题：检验 H₀: μ = 60 vs H₁: μ < 60，已知 σ = 8，n = 36，拒绝域为 x̄ < 57.8。
P(Type I error) = P(x̄ < 57.8 | μ = 60) = P(Z < (57.8−60)/(8/√36) = −2.2/1.333 = −1.65) = 0.0495
Now suppose the true mean is 58. Find the probability of Type II error and the power.
现假设真实均值为 58，计算第二类错误概率及功效。
P(Type II error) = P(x̄ ≥ 57.8 | μ = 58) = P(Z ≥ (57.8−58)/1.333 = −0.15) = P(Z ≥ −0.15) = 0.5596
Power = 1 − 0.5596 = 0.4404
Power curves can be generated by varying the true mean; exam questions may ask you to interpret such a curve.
变化真实均值可以得到功效曲线；考题可能要求解释该曲线的含义。
10. Common Mistakes and Top Tips | 常见错误与高分技巧
-
Forgetting that a PDF integral over the domain must equal 1 — many candidates lose easy marks by not solving for the constant correctly.
忘记概率密度函数在整个定义域上积分必须等于 1——许多考生因没有正确解出常数而丢分。
-
Treating continuous variables as discrete: P(X = x) is always zero for continuous distributions; always integrate to find probabilities.
将连续变量当作离散变量处理：连续分布中 P(X = x) 恒等于零，求概率必须积分。
-
Choosing the wrong critical value: one‑tailed vs two‑tailed
Published by TutorHao | A-Level Mathematics 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