📚 Investigation 4 - Probabilities of Compound Events | 探究四——复合事件的概率
Compound events lie at the heart of probability theory, blending multiple simple events into richer, more realistic models of uncertainty. This investigation guides you through the key rules governing unions, intersections, and conditional relationships, showing how seemingly separate outcomes connect and shape the chance of a combined result. By exploring additive and multiplicative paths both with and without replacement, you will build an intuition for dependence, independence, and the elegant mathematics that underpins real-world decision-making – from medical testing to risk analysis.
复合事件是概率论的核心，它将多个简单事件融合成更丰富、更贴近现实的不确定性模型。本次探究将带你领略支配并集、交集和条件关系的关键法则，展示看似独立的结果如何关联并影响组合结果的可能性。通过在有放回和无放回情况下探索加法与乘法路径，你将建立对相依性、独立性的直觉，以及支撑从医学检测到风险分析等现实决策的优雅数学。
1. From Simple to Compound: Defining the Landscape | 从简单到复合：界定领域
random experiment, such as rolling a 5 on a fair die. A compound event combines two or more simple events using operations like 'and' (intersection) or 'or' n number and a number greater than 4 on a single die is a compound event described by the intersection E ∩ rstanding how these building blocks link is the first step toward modelling layered uncertainty.
简单事件是随机实验的单一结果，例如在一枚均匀骰子上掷出5点。复合事件则通过“且”（交集）或“或”（并集）等运算将两个或多个简单事件组合起来。例如，一次掷骰既得偶数又大于4，这一复合事件可以用交集E ∩ G来描述，其中E = {2,4,6}，G = {5,6}。理解这些基本构件如何关联，是构建层级不确定性模型的第一步。
In set notation, the universal set S represents the sample space. The complement of an event A, written A' or Aᶜ, captures all outcomes not in A. Compound events allow us to ask richer questions: What is the chance of drawing a heart or a face card? What is the probability that two successive coin flips both land heads? These questions demand precise handling of the conceptual boundaries between union, intersection, and conditioning.
在集合符号中，全集S代表样本空间。事件A的补集，记作A'或Aᶜ，包含所有不在A中的结果。复合事件让我们能提出更丰富的问题：抽到红心或人头牌的概率是多少？连续两次抛硬币均正面的概率是多少？这些问题要求我们精确处理并集、交集和条件关系之间的概念边界。
2. The Addition Rule and Mutual Exclusivity | 加法法则与互斥性
ity of event A or event B occurring, we turn to the addition rule. For any two events, P(A ∪ B) = P(A) + P(B) − P(A ∩ ion prevents double-counting outcomes shared by A and B. If A and B are mutually exclusive – meaning they cannot occur together – then P(A ∩ B) = 0, and the rule simplifies to P(A ∪ B) = P(A) + P(B).
当我们求事件A或事件B发生的概率时，便可运用加法法则。对于任意两个事件，P(A ∪ B) = P(A) + P(B) − P(A ∩ B)。减去交集部分可避免对A和B共有结果的重复计算。若A与B互斥——即两者不可能同时发生——则P(A ∩ B) = 0，公式简化为P(A ∪ B) = P(A) + P(B)。
Imagine drawing a single card from a standard 52-card deck. Let A be 'drawing a King' and B be 'drawing a Queen' King and Queen. Hence, P(King or Queen) = 4/52 + 4/52 = 8/52 = 2/13. In contrast, if C is 'drawing a heart' and D is 'drawing a King', they are not mutually exclusive because the King of hearts belongs to both. Then P(C ∪ D) = 13/52 + 4/52 − 1/52 = 16/52 = 4/13.
设想从一副标准52张扑克牌中抽取一张。设A为“抽到老K”，B为“抽到皮蛋”。这两个事件互斥；一张牌不可能既是老K又是皮蛋。因此，P(老K或皮蛋) = 4/52 + 4/52 = 8/52 = 2/13。相反，若C为“抽到红心”，D为“抽到老K”，它们并非互斥，因为红心老K同时属于两者。此时P(C ∪ D) = 13/52 + 4/52 − 1/52 = 16/52 = 4/13。
3. The Multiplication Rule and Independent Events | 乘法法则与独立事件
utcome requires event A and event B to occur together, we use the multiplication rule: P(A ∩ B) = P(A) × P(B | A). Here P(B | ence of A does not affect the probability of B, the events are independent, satisfying P(B | A) = P(B). For independent events, the rule collapses to P(A ∩ B) = P(A) × P(B).
当一个复合结果要求事件A和事件B同时发生时，我们使用乘法法则：P(A ∩ B) = P(A) × P(B | A)。其中P(B | A)是已知A发生下B的条件概率。如果A的发生不影响B的概率，则两事件独立，满足P(B | A) = P(B)。对于独立事件，公式简化为P(A ∩ B) = P(A) × P(B)。
cally independent; knowledge of the coin outcome tells us nothing about the die. Thus, P(H ∩ 6) = ½ × ⅙ = 1/12 ≈ 0.0833. Independence is a powerful modelling assumption, but it must be justified – ependent events for independent ones is a common source of error in probability reasoning.
考虑抛一枚均匀硬币并掷一枚均匀骰子。硬币正面朝上（H）与骰子显示6点在物理上独立；获知硬币结果并不能告诉我们任何关于骰子的信息。因此，P(H ∩ 6) = ½ × ⅙ = 1/12 ≈ 0.0833。独立性是一个强大的建模假设，但必须有理由支撑——将相依事件误认为独立事件是概率推理中常见的错误源头。
4. Conditional Probability: Updating Beliefs | 条件概率：更新信念
The conditional probability of event A given event B is defined as P(A | B) = P(A ∩ he sample space from the universal set S down to B, re-scaling the overlap proportionally.
已知事件B发生下事件A的条件概率定义为P(A | B) = P(A ∩ B) / P(B)，前提是P(B) > 0。该公式刻画了当我们获知B已发生后A的概率如何改变。它实质上是将样本空间从全集S缩小至B，并按比例重新缩放重叠部分。
probability a randomly chosen student studies Physics given they study Mathematics is P(P | M) = P(M ∩ Mathematics raises the chance they also study Physics, indicating a positive association.
一个经典课堂示例：某组100名学生中，40人学习数学（M），30人学习物理（P），15人同时学习两科。随机选出一名学生，已知该生学习数学，则其也学习物理的概率为P(P | M) = P(M ∩ P) / P(M) = 15/40 = 0.375。若无条件，P(P) = 30/100 = 0.3；知道学生学数学提升了其也学物理的概率，表明两者存在正向关联。
5. Tree Diagrams and Sequential Experiments | 树形图与序贯实验
Tree diagrams efficiently organise multi-stage experiments, where outcomes at one stage feed into the next. Each branch is labelled with its conditional probability given the preceding path. The probability of a complete path is the product of the branch probabilities along it, directly applying the multiplication rule in a visual format. Tree diagrams also facilitate the calculation of total probabilities by summing the probabilities of relevant terminal nodes.
树形图能有效地组织多阶段实验，其中每一阶段的结果会汇入下一阶段。每条分支都标有给定前面路径的条件概率。一条完整路径的概率等于沿途分支概率的乘积，从而以直观形式直接应用乘法法则。树形图还有助于通过相加相关终端节点的概率来计算全概率。
= 0.6. For the second draw, the conditional probabilities shift: after a red first, P(Red₂ | Red₁) = 3/9 = 1/3; after a blue first, P(Red₂ | × 1/3 = 4/30 ≈ 0.1333. The tree makes dependencies explicit and prevents confusion between 'with replacement' and 'without replacement' scenarios.
假设一个袋子装有4颗红球和6颗蓝球。无放回地抽取两颗。第一阶段分支显示P(红₁) = 0.4，P(蓝₁) = 0.6。到第二次抽取时，条件概率发生变化：第一次红后，P(红₂ | 红₁) = 3/9 = 1/3；第一次蓝后，P(红₂ | 蓝₁) = 4/9。抽到两红的概率为P(RR) = 0.4 × 1/3 = 4/30 ≈ 0.1333。树形图清晰地展示了相依性，避免了“有放回”与“无放回”情景的混淆。
6. With Replacement versus Without Replacement | 有放回与无放回
The distinction between sampling with and without replacement fundamentally alters the dependence structure between draws. With replacement, successive trials are independent, and probabilities remain constant across stages. Without replacement, each draw changes the composition of the remaining population, making later draws dependent on earlier ones. Both settings appear frequently in IB investigations involving cards, marbles, and population sampling.
有放回抽样与无放回抽样的区别从根本上改变了各次抽取之间的相依结构。有放回时，各次试验相互独立，概率在各阶段保持不变。无放回时，每次抽取都会改变剩余总体的组成，使得后续抽取相依于先前结果。这两种设定在涉及纸牌、弹珠和总体抽样的IB探究中频繁出现。
The without-replacement probability is lower because the first Ace removed reduces the pool of Aces available for the second draw. Recognising which sampling model applies is crucial for correct probability calculations in real-world problems such as quality control inspections.
一个数值对比可阐明这种影响。从一副标准扑克牌中抽两张。有放回：P(两张均为A) = (4/52) × (4/52) = 16/2704 ≈ 0.00592。无放回：P(两张均为A) = (4/52) × (3/51) = 12/2652 ≈ 0.00452。无放回的概率较低，因为第一张A被移走后减少了第二张可用A的数量。识别所适用的抽样模型对于在如质量控制检验等现实问题中进行正确概率计算至关重要。
7. The Law of Total Probability | 全概率公式
onal probabilities across a set of mutually exclusive and exhaustive scenarios. If B₁, B₂, …, Bₙ form a partition of the sample space, then P(A) = Σ P(Bᵢ) × P(A | icult to evaluate but its conditional probabilities within known subgroups are accessible.
全概率公式将某一事件的概率表示为该事件在一组互斥且穷举的场景下条件概率的加权和。若B₁, B₂, …, Bₙ构成样本空间的一个划分，则P(A) = Σ P(Bᵢ) × P(A | Bᵢ)。当A的直接概率难以评估，但其在已知子组中的条件概率易于获取时，这一技巧极具价值。
Imagine a factory with two machines: Machine X produces 70% of items (defective rate 2%), Machine Y produces 30% (defective rate 5%). What is the probability a randomly selected item is defective? Let D be defective. P(D) = P(X)·P(D|X) + P(Y)·P(D| antly handles mixture problems and sets the stage for the inversion of conditions in Bayes' theorem.
设想一个工厂有两台机器：机器X生产70%的产品（次品率2%），机器Y生产30%（次品率5%）。随机抽取一件产品为次品的概率是多少？设D为次品。P(D) = P(X)·P(D|X) + P(Y)·P(D|Y) = 0.7 × 0.02 + 0.3 × 0.05 = 0.014 + 0.015 = 0.029。全概率公式优雅地处理了混合问题，并为贝叶斯定理中的条件反转奠定了基础。
8. Bayes' Theorem: Flipping the Condition | 贝叶斯定理：反转条件
Bayes' ing the probability of a hypothesis given new evidence. It reverses the conditioning: P(Bᵢ | A) = [P(Bᵢ) × P(A | . The theorem quantifies how prior beliefs (P(Bᵢ)) should shift to posterior beliefs (P(Bᵢ | A)) upon observing data A.
贝叶斯定理为根据新证据更新假设概率提供了数学框架。它反转了条件：P(Bᵢ | A) = [P(Bᵢ) × P(A | Bᵢ)] / P(A)，其中P(A)通过全概率公式求得。该定理量化了在观察到数据A后，先验信念（P(Bᵢ)）应如何转变为后验信念（P(Bᵢ | A)）。
given a randomly chosen item is defective, what is the probability it came from Machine Y? P(Y | D) = [P(Y) × P(D| (D) = (0.3 × 0.05) / 0.029 ≈ 0.015 / 0.029 ≈ 0.517. Even though Machine Y produces only 30% s, over half of the defective items originate from it due to its higher defect rate. Bayes' theorem reveals such counter-intuitive insights and is widely applied in medical screening, spam filtering, and machine learning.
延续工厂示例：已知随机抽取的一件产品为次品，它来自机器Y的概率是多少？P(Y | D) = [P(Y) × P(D|Y)] / P(D) = (0.3 × 0.05) / 0.029 ≈ 0.015 / 0.029 ≈ 0.517。尽管机器Y仅生产30%的产品，但由于其较高的次品率，超过一半的次品来自它。贝叶斯定理揭示了此类反直觉的洞见，并广泛应用于医学筛查、垃圾邮件过滤和机器学习中。
9. Venn Diagrams and Probability Tables | 文氏图与概率表
Visual representations such as Venn diagrams and two-way tables bring the abstract operations on compound events to life. A Venn diagram maps the sample space as a rectangle, with circles representing events A and B. The overlapping region represents the intersection A ∩ B, while the combined area of the circles represents the union A ∪ B. Shading specific regions solidifies understanding of complements and complex combinations like (A ∪ B)' or A ∩ B'.
文氏图和双向表等可视化表示将复合事件的抽象运算具象化。文氏图将样本空间绘作矩形，用圆表示事件A和B。重叠区域表示交集A ∩ B，而圆的合区表示并集A ∪ B。对特定区域进行着色可以巩固对补集以及诸如(A ∪ B)'或A ∩ B'等复杂组合的理解。
A two-way table (also called a contingency table) displays the frequency or probability distribution of two categorical variables. Consider gender (Male/Female) and subject preference (Math/Physics). The body of the table gives joint frequencies, while row and column totals give marginal frequencies. From such a table, one can read off P(Math ∩ Male), calculate row-conditional probabilities like P(Math | Female), and check for independence by comparing the product of marginals with the joint probability.
双向表（亦称列联表）展示两个分类变量的频数或概率分布。以性别（男/女）和学科偏好（数学/物理）为例。表体给出联合频数，行、列合计给出边缘频数。从这样的表中，我们可以直接读出P(数学 ∩ 男)，计算行条件概率如P(数学 | 女)，并通过比较边缘概率之积与联合概率来检验独立性。
10. Testing Independence in Data | 检验数据中的独立性
From a two-way table, events A and B are independent if and only if P(A ∩ he proportion of A within B equals the overall proportion of A. For example, if the probability a randomly selected student likes Math is the same regardless of gender, then preference and gender are independent. Slight deviations in real data are expected, but large discrepancies suggest an association.
从双向表来看，事件A与B独立当且仅当P(A ∩ B) = P(A) × P(B)，或者等价地，A在B内部的比例等于A的整体比例。例如，若无论性别如何，随机选出的学生喜欢数学的概率相同，则偏好与性别独立。在真实数据中，轻微偏差在所难免，但较大差距则提示存在关联。
70 boys and 55 girls enjoy coding. Overall P(coding) = 125/200 = 0.625. For boys, P(coding | nd enjoying coding in this sample. These comparisons form the basis of chi-squared tests for independence studied later in the IB syllabus.
考虑对200名学生的调查：90名男生，110名女生。70名男生和55名女生喜欢编程。整体P(编程) = 125/200 = 0.625。对于男生，P(编程 | 男) = 70/90 ≈ 0.778。条件概率明显更高，表明非独立，且在该样本中男生与喜欢编程之间存在正向关联。这类比较构成了IB课程后续学习的卡方独立性检验的基础。
11. Combined Strategies for Complex Problems | 复杂问题的综合策略
Many IB problems demand layered reasoning: identify the compound events, determine dependencies, draw a tree or complete a table, apply addition/multiplication rules, and perhaps use total probability to link branches. Worked example: In a game, a player rolls a die and then draws a card from a bag containing 5 red and 3 green cards if the roll is even; otherwise draws from a bag with 2 red and 6 green. Find the probability of drawing a red card.
许多IB问题需要分层推理：识别复合事件，确定相依关系，绘制树形图或填写表格，应用加法/乘法法则，或许还要运用全概率公式连接分支。详解示例：某游戏中，玩家掷一枚骰子，若点数为偶数则从装有5红3绿的袋中抽卡；否则从装有2红6绿的袋中抽卡。求抽到红卡的概率。
Step 1: P(Even) = ½, P(Odd) = ½. Step 2: P(Red | Even) = 5/8, P(Red | l of reducing everything to a single formula prematurely. Always draw the decision structure first, then assign probabilities.
步骤1：P(偶数) = ½，P(奇数) = ½。步骤2：P(红 | 偶数) = 5/8，P(红 | 奇数) = 2/8 = ¼。步骤3：由全概率公式，P(红) = ½ × 5/8 + ½ × 2/8 = 5/16 + 2/16 = 7/16。以结构化的方式组合运用法则，能避免过早将所有东西简化为单一公式这一常见陷阱。务必先画出决策结构，再赋值概率。
12. Investigating Real-World Misconceptions | 探究现实中的误区
Compound probability is fertile ground for cognitive biases. The gambler's fallacy – the belief that a run of one outcome makes the opposite outcome 'due' – arises from misapplying the law of large numbers to short sequences. In independent trials, past outcomes do not alter the probabilities of future events. If a fair coin lands heads five times in a row, P(heads on sixth toss) remains ½. Investigating such fallacies with simulated data deepens critical thinking.
复合概率是认知偏见的沃土。赌徒谬误——认为一连串的某种结果会使相反结果“该来了”——源于将大数定律误用到短序列上。在独立试验中，已发生的结果不会改变未来事件的概率。若一枚均匀硬币连续五次正面朝上，第六次抛掷的P(正面)依然是½。通过模拟数据探究这类谬误能深化批判性思维。
e the conjunction of two events as more probable than one of the events alone, defying P(A ∩ B) ≤ P(A). For instance, in the famous Linda problem, many rate 'feminist bank teller' as more likely than 'bank teller' ion of such phenomena connects mathematical probability to psychology and human reasoning.
另一个常见错误是合取谬误，即人们错误地认为两个事件的合取比单独一个事件更可能发生，违背了P(A ∩ B) ≤ P(A)。例如，在著名的琳达问题中，许多人认为“女权主义银行柜员”比“银行柜员”更可能，忽略了合取在逻辑上不可能大于其组成部分。对此类现象的讨论将数学概率与心理学及人类推理联系起来。
Published by TutorHao | IB 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