Edexcel AS & A level Further Statistics 1: Core Topics Overview | Edexcel AS与A level进阶统计1 核心主题总览

📚 Edexcel AS & A level Further Statistics 1: Core Topics Overview | Edexcel AS与A level进阶统计1 核心主题总览

Further Statistics 1 (FS1) is a key module in the Edexcel AS and A level Further Mathematics course, designed to deepen your understanding of probability distributions, statistical inference, and hypothesis testing. It extends ideas from the pure Mathematics and Statistics sections of A level Mathematics, introducing new discrete and continuous distributions, advanced hypothesis tests such as the chi-squared test, and the Central Limit Theorem. Mastering this module requires fluency in algebraic manipulation, summation notation, and calculator-based computation for critical values and probabilities.

进阶统计1（FS1）是Edexcel AS与A level进阶数学的核心模块，旨在加深你对概率分布、统计推断与假设检验的理解。它从A level数学的纯数与统计部分延伸出来，引入新的离散与连续分布、卡方检验等高级假设检验以及中心极限定理。掌握这个模块需要熟练的代数运算、求和符号运用，并能用计算器求出临界值和概率。

---

1. Discrete Random Variables and Expectation | 离散随机变量与期望

ariable X, the expected value is E(X) = Σ x P(X = x). The variance is defined as Var(X) = E(X²) – combinations also features prominently: E(aX + b) = aE(X) + b and Var(aX + b) = a² Var(X).

在FS1中，我们重温离散随机变量，并更注重从基本原理推导均值与方差。对于任一离散随机变量X，期望值为E(X) = Σ x P(X = x)。方差定义为Var(X) = E(X²) – [E(X)]²，其中E(X²) = Σ x² P(X = x)。这些求和覆盖X的所有可能取值。期望与方差的线性性质在线性组合中也十分突出：E(aX + b) = aE(X) + b，Var(aX + b) = a² Var(X)。

r(X + Y) = Var(X) + Var(Y). These properties underpin nearly all subsequent topics in FS1.

处理独立随机变量X与Y时，和的均值为E(X + Y) = E(X) + E(Y)，和的方差为Var(X + Y) = Var(X) + Var(Y)。这些性质奠定了FS1几乎全部后续内容的基础。

---

2. Poisson Distribution | 泊松分布

me or space, assuming events happen independently at a constant average rate λ. We write X ~ Po(λ). The probability mass function is P(X = r) = e⁻λ λʳ / r!, for r = 0, 1, 2, … events, such as the number of calls to a helpline per minute or flaws per metre of fabric.

泊松分布用于对固定时间或空间区间内发生的事件数量建模，假设事件独立发生且具有恒定的平均发生率λ。记作X ~ Po(λ)。其概率质量函数为P(X = r) = e⁻λ λʳ / r!，其中r = 0, 1, 2, … 。均值与方差都等于λ。泊松分布常被用于稀有事件，例如每分钟打入求助热线的电话数或每米布料上的瑕疵数。

is approximation is acceptable when n > 50 and np < 5. The sum of independent Poisson variables is also Poisson: if X ~ Po(λ₁) and Y ~ Po(λ₂), then X + Y ~ Po(λ₁ + λ₂).

一个重要的条件是当二项分布B(n, p)的n很大而p很小时，可用泊松分布近似，令λ = np。通常n > 50且np < 5时近似效果可接受。独立泊松变量之和仍服从泊松分布：若X ~ Po(λ₁)且Y ~ Po(λ₂)，则X + Y ~ Po(λ₁ + λ₂)。

---

3. Geometric Distribution | 几何分布

th success probability p. Two forms exist in Edexcel: FS1 uses the version where the random variable X represents the trial number on which the first success occurs, so that P(X = r) = p (1 − p)ʳ⁻¹, for r = 1, 2, 3, … . The mean is E(X) = 1/p and the variance is Var(X) = (1 − p) / p².

几何分布对在一系列独立的伯努利试验中直到（且包括）首次成功所需的试验次数进行建模。Edexcel采用的形式是随机变量X代表首次成功发生的试验序号，因此P(X = r) = p (1 − p)ʳ⁻¹, r = 1, 2, 3, … 。期望为E(X) = 1/p，方差为Var(X) = (1 − p) / p²。

he same as starting afresh. This property can be tested in exam questions, often by checking whether P(X > r + s | X > r) = P(X > s).

几何分布具有无记忆性：已知试验r次尚未成功，则首次成功发生在r + s次以后的概率与重新开始一样。考卷中可能会检验这一性质，通常验证P(X > r + s | X > r) = P(X > s)。

---

4. Negative Binomial Distribution | 负二项分布

r. If X denotes the trial on which the r-th success occurs, then P(X = x) = C(x−1, r−1) pʳ (1 − p)ˣ⁻ʳ, for x = r, r+1, … ritten with combinations. The mean is E(X) = r / p and variance is Var(X) = r(1 − p) / p².

负二项分布将几何模型扩展到获得固定次数成功r所需的试验次数。如果X表示第r次成功发生的试验序号，则P(X = x) = C(x−1, r−1) pʳ (1 − p)ˣ⁻ʳ, x = r, r+1, … 。在FS1中常使用组合记号。期望为E(X) = r / p，方差为Var(X) = r(1 − p) / p²。

You can also view the negative binomial variable Y as the number of failures before the r-th success, but the Edexcel FS1 specification treats count of trials. Sums of independent geometric random variables with the same p form a negative binomial: the sum of r independent Geo(p) variables follows NB(r, p) for the trial count.

你也可以将负二项变量Y看作第r次成功之前的失败次数，但Edexcel FS1大纲采用试验次数。具有相同p的独立几何随机变量之和形成负二项分布：r个独立的Geo(p)变量的和服从NB(r, p)（试验计数版本）。

---

广告 / Sponsored

5. Continuous Random Variables and PDFs | 连续随机变量与概率密度函数

rea under the curve equals 1, i.e. ∫ f(x) dx = 1 over the domain. The probability that X lies in an interval [a, b] is given by the definite integral P(a ≤ X ≤ b) = ∫ₐᵇ f(x) dx.

FS1通过概率密度函数f(x)引入连续随机变量。pdf必须对所有x满足f(x) ≥ 0，且曲线下的总面积为1，即在其定义域上∫ f(x) dx = 1。X落在区间[a, b]内的概率由定积分P(a ≤ X ≤ b) = ∫ₐᵇ f(x) dx给出。

The cumulative distribution function (cdf) is F(x) = P(X ≤ x) = ∫₋ₒₒˣ c. The mean and variance are defined as E(X) = ∫ x f(x) dx and Var(X) = ∫ x² f(x) dx − μ².

累积分布函数为F(x) = P(X ≤ x) = ∫₋ₒₒˣ f(t) dt。中位数、四分位数和百分位数可通过求解F(x) = 0.5, 0.25, 0.75等得到。均值与方差定义为E(X) = ∫ x f(x) dx，Var(X) = ∫ x² f(x) dx − μ²。

---

6. Continuous Uniform Distribution | 连续均匀分布

b − a) for a ≤ x ≤ b, and 0 otherwise. The mean is (a + b)/2, and the variance is (b − a)²/12. The cumulative distribution function grows linearly: F(x) = (x − a)/(b − a) for x in [a, b].

FS1中最简单的连续分布是区间[a, b]上的连续均匀分布，记作U(a, b)。其概率密度函数为f(x) = 1/(b − a) (a ≤ x ≤ b)，否则为0。均值为(a + b)/2，方差为(b − a)²/12。累积分布函数线性增长：F(x) = (x − a)/(b − a)，当x ∈ [a, b]。

Exam problems often require finding conditional probabilities or percentiles. For example, P(X > c | X > d) can be computed directly from the c.d.f. because of the simple form. Additionally, rectangular distributions are used to model rounding errors or waiting times when arrivals are uniformly random.

考题常要求计算条件概率或百分位数。例如，由于形式简单，P(X > c | X > d) 可直接从c.d.f.算出。此外，矩形分布可用于模拟舍入误差或当到达时间为完全随机时的等待时间。

---

7. Hypothesis Testing with Discrete Distributions | 离散分布的假设检验

ull hypothesis H₀ (e.g. λ = 5) and alternative hypothesis H₁ (λ > 5, λ < 5, or λ ≠ 5). Using the test statistic and observed data, you calculate the probability of obtaining a result at least as extreme as the observed one, assuming H₀ is true. This p-value is compared to the significance level α (commonly 5% or 1%).

FS1规范了离散分布（泊松、二项、几何、负二项）参数的假设检验流程。设定原假设H₀（如λ = 5）与备择假设H₁（λ > 5，λ < 5 或 λ ≠ 5）。利用检验统计量与观测数据，计算在原假设成立下得到至少与观测结果一样极端的概率。此p值与显著性水平α（通常为5%或1%）比较。

₁: λ > λ₀, the critical region is X ≥ c, where c is the smallest integer such that P(X ≥ c | λ₀) ≤ α. For two-tailed tests, the significance level is shared between both tails.

也可采用临界区域法：由使H₀被拒绝的值构成的集合。对于泊松检验且H₁: λ > λ₀，临界区域为X ≥ c，其中c是满足P(X ≥ c | λ₀) ≤ α的最小整数。对于双尾检验，显著性水平被分配在两个尾部。

---

8. Chi-Squared Goodness-of-Fit Test | 卡方拟合优度检验

Eᵢ)² / Eᵢ, where Oᵢ are observed frequencies and Eᵢ are expected frequencies under the model. Under H₀, this statistic approximately follows a χ² distribution with ν degrees of freedom.

卡方拟合优度检验可检验观测频数是否与理论模型相符。检验统计量为X² = Σ (Oᵢ − Eᵢ)² / Eᵢ，其中Oᵢ为观测频数，Eᵢ为该模型下的期望频数。在H₀下，此统计量近似服从自由度为ν的χ²分布。

ories − 1 − p, where p is the number of parameters estimated from the data. You must ensure all expected frequencies are at least 5; adjacent categories may be combined if necessary. The critical value is obtained from χ² tables; reject H₀ if X² exceeds this value.

自由度ν = 分类数 − 1 − p，其中p是从数据中估计的参数个数。必须保证所有期望频数至少为5；必要时可合并相邻类别。从χ²表中查得临界值；若X²超出临界值则拒绝H₀。

---

9. Chi-Squared Test for Independence | 独立性卡方检验

iables are associated. Expected frequencies are calculated as E = (row total × column total) / grand total. Degrees of freedom for an m × n table are (m − 1)(n − 1). The test statistic is the same sum: Σ (O − E)² / E.

当数据按两个分类变量排成一个列联表时，独立性卡方检验可检测变量是否有关联。期望频数按E = (行合计 × 列合计) / 总计计算。m × n表的自由度为(m − 1)(n − 1)。检验统计量仍是求和：Σ (O − E)² / E。

ives a significant result, you reject the null hypothesis of independence and conclude there is evidence of an association. You are not expected to measure the strength, only to test.

同样，近似要求期望频数≥ 5。若检验结果显著，则拒绝独立性的原假设，得出结论“存在关联的证据”。你无需衡量关联强度，仅需进行检验。

---

10. Central Limit Theorem | 中心极限定理

ndom sample of size n from any distribution with mean μ and variance σ², the sample mean X̄ ypically n ≥ 30). This justifies using normal approximations in many inference procedures.

中心极限定理指出，对来自均值为μ、方差为σ²的任意总体的容量为n的随机样本，当n足够大（通常n ≥ 30）时，样本均值X̄近似服从均值为μ、方差为σ²/n的正态分布。这使得在许多推断过程中可以使用正态近似。

Applications in FS1 include approximating the sum of i.i.d. discrete random variables, finding probabilities for sample means, and solving problems where the original distribution is unknown. The continuity correction is not required for the sample mean when using the CLT directly, but you must be careful when approximating a discrete sum.

FS1中的应用包括近似独立同分布离散随机变量的总和、求样本均值的概率以及解决原分布未知的问题。直接使用CLT时，样本均值无需进行连续性校正，但在近似离散总和时需谨慎。

Published by TutorHao | Further Statistics 1 Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）