📚 Edexcel AS and A Level Further Mathematics: Further Statistics 2 Key Topics Explained | Edexcel AS与A Level进阶数学：进阶统计学2知识点精讲
Further Statistics 2 (FS2) builds on earlier probability and statistics units, introducing rigorous techniques for continuous random variables, estimation, hypothesis testing, chi-squared tests and probability generating functions. Mastering these topics is essential for deep statistical analysis and success in the A Level exam.
进阶统计学2 (FS2) 在前期概率与统计单元的基础上，引入了连续随机变量、估计、假设检验、卡方检验以及概率生成函数等严谨技巧。掌握这些专题对于深入统计分析并在 A Level 考试中取得优异成绩至关重要。
1. Continuous Random Variables and Probability Density Functions | 连续随机变量与概率密度函数
DF), f(x), which must satisfy f(x) ≥ 0 for all x, and the total area under the curve equals 1: ∫-∞∞ f(x) dx = 1.
连续随机变量 X 可在某区间内取任意值。其概率分布由概率密度函数 (PDF) f(x) 定义，该函数必须满足对所有 x 有 f(x) ≥ 0，且曲线下总面积为 1：∫-∞∞ f(x) dx = 1。
The probability that X lies between a and b is given by the integral P(a < X < b) = ∫ab f(x) dx Since continuous variables have no probability at a single point, P(X = c) = 0 for any c.
X 落在 a 与 b 之间的概率由积分 P(a < X < b) = ∫ab f(x) dx 给出。由于连续变量在单点处概率为零，因此对任意 c 有 P(X = c) = 0。
The PDF is typically defined piecewise, and its shape determines how probabilities are distributed. Common examples include the uniform, exponential and normal PDFs.
概率密度函数通常分段定义，其形状决定了概率的分布方式。常见例子包括均匀分布、指数分布和正态分布的 PDF。
2. Cumulative Distribution Functions and Percentiles | 累积分布函数与百分位数
The cumulative distribution function (CDF), F(x), gives the probability that X is less than or equal to x: F(x) = P(X ≤ x) = ∫-∞x f(t) dt. It is a non-decreasing function with limits F(-∞) = 0 and F(∞) = 1.
累积分布函数 (CDF) F(x) 给出 X 小于等于 x 的概率：F(x) = P(X ≤ x) = ∫-∞x f(t) dt。它是一个非递减函数，极限为 F(-∞) = 0 和 F(∞) = 1。
The lower quartile Q₁ is found from F(Q₁)=0.25 and the upper quartile Q₃ from F(Q₃)=0.75.
包括中位数在内的百分位数通过求解 F(x) = p 得到。例如，中位数 m 满足 F(m) = 0.5。下四分位数 Q₁ 由 F(Q₁) = 0.25 求得，上四分位数 Q₃ 由 F(Q₃) = 0.75 求得。
When the CDF is given in piecewise form, locate the correct interval and solve for the unknown x. Always verify that the solution lies within the interval where the equation is valid.
当 CDF 为分段形式时，需要找到正确的区间并求解未知数 x。务必验证解落在该方程适用的区间内。
3. Expectation and Variance of Continuous Distributions | 连续分布的期望与方差
The expectation (mean) of a continuous random variable X is E(X) = ∫ x f(x) dx over the entire range. For a function g(X), E(g(X)) = ∫ g(x) f(x) dx.
连续随机变量 X 的期望（均值）为整个范围内的积分 E(X) = ∫ x f(x) dx。对于函数 g(X)，E(g(X)) = ∫ g(x) f(x) dx。
The variance is defined by Var(X) = E[(X - μ)²] = ∫ (x - μ)² f(x) dx, and can be computed more easily using Var(X) = E(X²) - [E(X)]². The standard deviation is σ = √Var(X).
方差定义为 Var(X) = E[(X - μ)²] = ∫ (x - μ)² f(x) dx，也可更简便地使用 Var(X) = E(X²) - [E(X)]² 计算。标准差为 σ = √Var(X)。
These integrals are usually evaluated over the domain where f(x) > 0. For symmetric PDFs, the mean may be quickly identified; for other distributions, integration by parts or substitution may be required.
这些积分通常在 f(x) > 0 的区域内计算。对于对称 PDF，均值可快速识别；对于其他分布，可能需要使用分部积分或代换法。
4. Linear Combinations of Independent Random Variables | 独立随机变量的线性组合
If X and Y are independent random variables with known means and variances, then for constants a and b, E(aX + bY) = aE(X) + bE(Y) and Var(aX + bY) = a²Var(X) + b²Var(Y). The result extends to sums of more than two variables.
若 X 和 Y 为独立随机变量且均值与方差已知，则对于常数 a 和 b，有 E(aX + bY) = aE(X) + bE(Y) 以及 Var(aX + bY) = a²Var(X) + b²Var(Y)。该结论可推广至多于两个变量的和。
d variance σ², the sample total T = Σ Xᵢ has E(T) = nμ and Var(T) = nσ². The sample mean X̄ = T/n then has E(X̄) = μ and Var(X̄) = σ²/n.
特别地，对于独立同分布 (i.i.d.) 随机变量 X₁, X₂, ..., Xₙ，每个均值为 μ、方差为 σ²，样本总和 T = Σ Xᵢ 满足 E(T) = nμ、Var(T) = nσ²。样本均值 X̄ = T/n 则满足 E(X̄) = μ、Var(X̄) = σ²/n。
If the variables are normally distributed, any linear combination is also normally distributed. This property underpins many inference procedures in FS2.
若变量服从正态分布，则任意线性组合也服从正态分布。这一性质是 FS2 中许多推断方法的基础。
5. Unbiased Estimators and Sample Mean | 无偏估计量与样本均值
An estimator θ̂ is unbiased for a parameter θ if E(θ̂) = θ. The sample mean X̄ is an unbiased estimator of the population mean μ because E(X̄) = μ.
若估计量 θ̂ 满足 E(θ̂) = θ，则称其为参数 θ 的无偏估计量。样本均值 X̄ 是总体均值 μ 的无偏估计量，因为 E(X̄) = μ。
The sample variance S² = (1/(n-1)) Σ (Xᵢ - X̄ with denominator n is biased. Understanding bias helps in choosing appropriate estimators.
样本方差 S² = (1/(n-1)) Σ (Xᵢ - X̄)² 是总体方差 σ² 的无偏估计量，而分母为 n 的朴素形式则是有偏的。理解偏差有助于选择合适的估计量。
The Central Limit Theorem (CLT) states that for large n, the distribution of X̄ is approximately normal with mean μ and variance σ²/n, regardless of the shape of the original population, provided the variance is finite.
中心极限定理 (CLT) 指出，当 n 较大时，不论原始总体的形态如何，只要方差有限，X̄ 的分布近似正态，均值为 μ、方差为 σ²/n。
6. Confidence Intervals for the Mean (Variance Known) | 均值的置信区间（方差已知）
When a population is normally distributed with known variance σ², a (1-α)100% confidence interval for μ is given by x̄ ± zα/2 × (σ / √n), where zα/2 is the critical value from the standard normal distribution.
当总体服从正态分布且方差 σ² 已知时，μ 的 (1-α)100% 置信区间为 x̄ ± zα/2 × (σ / √n)，其中 zα/2 是标准正态分布的临界值。
For a 95% confidence interval, z0.025 eases as the sample size n increases, reflecting greater precision. Always interpret the interval correctly: we are 95% confident that the interval contains the true population mean μ.
对于 95% 置信区间，z0.025 = 1.96。区间宽度随样本量 n 增大而减小，反映出更高的精度。务必正确解读区间：我们有 95% 的信心认为该区间包含了真实的总体均值 μ。
If the population variance is unknown but the sample is large, σ can be replaced by the sample standard deviation s, and the interval remains approximately valid.
若总体方差未知但样本较大，可用样本标准差 s 替代 σ，此时区间近似有效。
7. Hypothesis Testing for the Mean Using Normal Distribution | 正态分布均值的假设检验
To test a hypothesis about a population mean μ with known variance σ², we use the test statistic Z = (x̄ - μ₀) / (σ / √n). Under the null hypothesis H₀: μ = μ₀, Z ~ N(0,1).
要检验已知方差 σ² 的总体均值 μ，我们使用检验统计量 Z = (x̄ - μ₀) / (σ / √n)。在零假设 H₀: μ = μ₀ 下，Z ~ N(0,1)。
For a two-tailed test H₁: μ ≠ μ₀, we reject H₀ if |Z| > zα/2. For one-tailed tests, compare Z to zα (upper-tail) or -zα (lower-tail). Alternatively, use p-values: if p < α, reject H₀.
对于双侧检验 H₁: μ ≠ μ₀，若 |Z| > zα/2 则拒绝 H₀。对于单侧检验，将 Z 与 zα（上尾）或 -zα（下尾）比较。也可用 p 值判决：若 p < α 则拒绝 H₀。
Always state a conclusion in context, and never use phrases like 'accept H₀'; instead, say 'there is insufficient evidence to reject H₀'.
始终在上下文中给出结论，切勿使用“接受 H₀”等表述；应该说“没有足够证据拒绝 H₀”。
8. Chi-Squared Goodness-of-Fit Test | 卡方拟合优度检验
The chi-squared goodness-of-fit test checks whether observed frequencies fit a specified distribution. The test statistic is χ² = Σ (Oᵢ - Eᵢ)² / Eᵢ, where Oᵢ are observed frequencies and Eᵢ are expected frequencies under the null hypothesis.
卡方拟合优度检验用于检验观测频数是否符合特定分布。检验统计量为 χ² = Σ (Oᵢ - Eᵢ)² / Eᵢ，其中 Oᵢ 为观测频数，Eᵢ 为零假设下的期望频数。
bability for category i under the theoretical model. The test is valid only if all expected frequencies are at least 5 (or a small proportion is less than 5).
期望频数通过 Eᵢ = npᵢ 计算，其中 pᵢ 是理论模型下第 i 类的概率。只有当所有期望频数都至少为 5（或仅有极小比例低于 5）时，检验才有效。
here k is the number of categories and m is the number of parameters estimated from the data. Compare χ² to a critical value from χ²-distribution with ν degrees of freedom.
自由度为 ν = k - 1 - m，其中 k 为类别数，m 为从数据中估计的参数个数。将 χ² 与自由度为 ν 的 χ² 分布临界值比较。
9. Chi-Squared Test for Independence (Contingency Tables) | 独立性卡方检验（列联表）
A contingency table displays frequencies for two categorical variables. The chi-squared test for independence assesses whether the variables are associated. The test statistic is again χ² = Σ (O - E)² / E, summed over all cells.
列联表展示了两个分类变量的频数。独立性卡方检验用于评估变量是否有关联。检验统计量同样是 χ² = Σ (O - E)² / E，对所有单元格求和。
Expected frequencies for each cell are calculated as E = (row total × column total) / grand total rees of freedom are ν = (r - 1)(c - 1), where r and c are the numbers of rows and columns.
每个单元格的期望频数计算为 E = (行总和 × 列总和) / 总频数。自由度 ν = (r - 1)(c - 1)，其中 r 和 c 分别为行数和列数。
A significant result suggests that the variables are not independent. When interpreting results, examine the contributions to χ² to identify cells driving the association.
显著的结果表明变量间不独立。在解释结果时，要检查各单元格对 χ² 的贡献，以识别导致关联的那些单元格。
10. Probability Generating Functions (PGFs) | 概率生成函数
For a discrete random variable X taking non-negative integer values, the probability generating function (PGF) is defined as GX(t) = E(tX) = Σ tˣ P(X = x), where the sum is over all possible x.
对于取非负整数值的离散随机变量 X，概率生成函数 (PGF) 定义为 GX(t) = E(tX) = Σ tˣ P(X = x)，其中求和涵盖所有可能的 x。
Key properties: GX(1) = 1. The first derivative gives the mean: G′X(1) = E(X). The second derivative helps find variance: Var(X) = G″X(1) + G′X(1) - [G′X(1)]².
关键性质：GX(1) = 1。一阶导数给出均值：G′X(1) = E(X)。二阶导数用于求方差：Var(X) = G″X(1) + G′X(1) - [G′X Published by TutorHao | A-Level Mathematics Revision Series | aleveler.com
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