Poisson Distribution: Basic Concepts and Characteristics | 泊松分布的基本概念与特征

📚 Poisson Distribution: Basic Concepts and Characteristics | 泊松分布的基本概念与特征

In A-Level Further Mathematics (Edexcel), the Poisson distribution is one of the most important discrete probability distributions. It models the number of times an event occurs in a fixed interval of time or space, when events happen independently and at a constant average rate. Understanding its definition, key properties, and underlying assumptions is essential for solving exam problems with confidence.

在 Edexcel A-Level 进阶数学中，泊松分布是最重要的离散概率分布之一。它用于建模在固定时间或空间区间内某事件发生的次数，前提是事件独立发生且平均速率恒定。理解其定义、关键性质与基本假设，是自信应对考试题目的基础。

1. Definition of the Poisson Distribution | 泊松分布的定义

A discrete random variable X is said to follow a Poisson distribution with parameter λ (lambda), written as X ~ Po(λ), if its probability mass function is given by:

离散型随机变量 X 服从参数为 λ（lambda）的泊松分布，记作 X ~ Po(λ)，其概率质量函数为：

P(X = r) = e⁻λ × λʳ / r! for r = 0, 1, 2, 3, …

Here, e is Euler’s number (approximately 2.71828), λ is the average rate of occurrence (which must be positive), and r! denotes the factorial of r. For example, 4! = 4 × 3 × 2 × 1 = 24, and by convention 0! = 1.

其中 e 是欧拉数（约等于 2.71828），λ 是事件发生的平均速率（必须为正数），r! 表示 r 的阶乘。例如，4! = 4 × 3 × 2 × 1 = 24，按惯例规定 0! = 1。

Note that X can take infinitely many non-negative integer values: 0, 1, 2, 3, and so on. Unlike the binomial distribution which has a finite number of outcomes, the Poisson distribution has an infinite range, although probabilities become extremely small for very large values of r.

注意，X 可以取无限多个非负整数值：0, 1, 2, 3 等等。与具有有限个结果的二项分布不同，泊松分布的取值范围是无限的，尽管对于非常大的 r 值，概率会变得极其微小。

To verify that the probabilities sum to 1, we use the series expansion eλ = Σ λʳ / r!. Substituting the Poisson formula gives Σ P(X = r) = e⁻λ × eλ = 1. This confirms that the Poisson distribution is a valid probability distribution.

为验证所有概率之和等于 1，我们使用级数展开 eλ = Σ λʳ / r!，代入泊松公式得 Σ P(X = r) = e⁻λ × eλ = 1。这确认了泊松分布是一个合法的概率分布。

2. The Meaning of the Parameter λ | 参数 λ 的含义

The parameter λ represents both the mean and the variance of the Poisson distribution. More precisely, if X ~ Po(λ), then E(X) = λ and Var(X) = λ. This remarkable property — that the mean equals the variance — distinguishes the Poisson distribution from most other distributions.

参数 λ 同时代表泊松分布的均值与方差。更确切地说，若 X ~ Po(λ)，则 E(X) = λ 且 Var(X) = λ。这一显著性质——即均值等于方差——使泊松分布区别于大多数其他分布。

In practical terms, λ is the average number of occurrences in the given interval. For example, if a call centre receives an average of 5 calls per minute, then for a one-minute interval we would model the number of calls as X ~ Po(5). If we instead consider a 10-minute interval, the new parameter would be λ = 5 × 10 = 50.

在实际问题中，λ 是给定区间内发生次数的平均值。例如，若某呼叫中心平均每分钟接到 5 通电话，那么在一分钟区间内我们可以将通话次数建模为 X ~ Po(5)。若改为考虑 10 分钟区间，则新参数为 λ = 5 × 10 = 50。

This proportionality of λ with the length of the interval is fundamental. When solving problems, always ensure that λ is scaled to match the interval or region specified in the question.

λ 与区间长度的比例关系是根本性的。解题时务必确保 λ 已被调整到与题目所给的区间或区域相匹配。

3. Key Assumptions of the Poisson Distribution | 泊松分布的关键假设

Before applying the Poisson distribution, it is critical to verify that the underlying situation satisfies certain assumptions. In Edexcel exam questions, these assumptions are often explicitly stated or need to be inferred from the context.

在应用泊松分布之前，必须验证实际问题是否满足某些假设。在 Edexcel 考试题中，这些假设常常被直接给出，或需要从题目情境中推断出来。

The first assumption is that events occur independently. This means that the occurrence of one event does not affect the probability of another event occurring. For example, the number of radioactive decays in a sample should not influence subsequent decays.

第一个假设是事件独立发生。这意味着一个事件的发生不会影响另一个事件发生的概率。例如，样本中放射性衰变的次数不应影响后续的衰变事件。

The second assumption is that events occur at a constant average rate. The expected number of occurrences in an interval is proportional to the length of that interval, but the actual number fluctuates randomly around this mean.

第二个假设是事件以恒定平均速率发生。区间内事件的期望个数与区间长度成正比，但实际次数在均值附近随机波动。

The third assumption is that two or more events cannot occur at exactly the same instant. In a sufficiently small time interval, the probability of more than one event is negligible. Mathematically, as the interval length tends to zero, the probability of two or more events tends to zero faster than the interval length itself.

第三个假设是两个或多个事件不可能在同一瞬间发生。在足够小的时间区间内，发生多于一个事件的概率可以忽略不计。从数学上讲，当区间长度趋于零时，两个或更多事件发生的概率比区间长度本身更快地趋于零。

If these assumptions are violated, the Poisson model may not be appropriate. For example, if events tend to cluster (such as earthquakes of large magnitude), independence fails and an alternative model should be considered.

如果这些假设不成立，泊松模型可能就不适用。例如，如果事件趋于聚集（如大地震），独立性就不成立，此时应考虑其他模型。

4. Mean and Variance of the Poisson Distribution | 泊松分布的均值与方差

For X ~ Po(λ), the expected value is derived directly from the definition:

对于 X ~ Po(λ)，期望值可以直接从定义推导得出：

E(X) = Σ r × e⁻λ λʳ / r! = λ

The derivation involves recognising that the sum reduces to λ multiplied by the sum of probabilities of a Po(λ) distribution. The variance calculation is similar and requires E(X²), which ultimately gives Var(X) = E(X²) − [E(X)]² = λ² + λ − λ² = λ.

推导过程利用了对 Po(λ) 分布求和后结果简化为 λ 的性质。方差的计算类似，需要先求出 E(X²)，最终得到 Var(X) = E(X²) − [E(X)]² = λ² + λ − λ² = λ。

Therefore, the mean and variance are both exactly equal to λ. This mean-variance equality is a powerful diagnostic tool: if a data set has a sample mean very different from its sample variance, a Poisson model is probably inappropriate.

因此，均值和方差都精确等于 λ。这种均值-方差相等是强大的诊断工具：如果一组数据的样本均值与样本方差差别很大，那么泊松模型很可能不适用。

It is also worth noting that the standard deviation is √λ. For a Poisson distribution with a large λ, approximately 95% of the probability lies within 2 standard deviations of the mean, i.e., within λ ± 2√λ.

还需注意，标准差为 √λ。对于 λ 较大的泊松分布，约 95% 的概率位于均值两侧各 2 个标准差的范围内，即在 λ ± 2√λ 之内。

5. Shape and Behaviour of the Distribution | 分布的形状与行为

When λ is small, the Poisson distribution is heavily skewed to the right. The most probable value is 0 or a small number, and probabilities tail off gradually. For example, when λ = 0.5, P(X = 0) = e⁻⁰·⁵ ≈ 0.607, which is the highest single probability.

当 λ 较小时，泊松分布高度右偏。最可能的值是 0 或较小的数，概率逐渐衰减。例如，当 λ = 0.5 时，P(X = 0) = e⁻⁰·⁵ ≈ 0.607，这是最大的单点概率。

As λ increases, the distribution becomes more symmetric and bell-shaped. When λ is around 20 or larger, the Poisson distribution closely resembles a normal distribution with mean λ and variance λ. This is a consequence of the Central Limit Theorem applied to independent Poisson counts.

随着 λ 增大，分布变得更对称、更呈钟形。当 λ 约为 20 或更大时，泊松分布十分接近均值为 λ、方差为 λ 的正态分布。这是中心极限定理应用于独立泊松计数的一个结果。

The mode (most probable value) of a Poisson distribution is either ⌊λ⌋ or ⌊λ⌋ − 1, where ⌊λ⌋ is the integer part of λ. In fact, the mode is the largest integer r such that P(X = r) ≥ P(X = r − 1), which simplifies to r ≤ λ. This means the mode is ⌊λ⌋ if λ is not an integer, and both λ and λ − 1 if λ is an integer.

泊松分布的众数（最可能的值）是 ⌊λ⌋ 或 ⌊λ⌋ − 1，其中 ⌊λ⌋ 表示 λ 的整数部分。事实上，众数是满足 P(X = r) ≥ P(X = r − 1) 的最大整数 r，化简得 r ≤ λ。这意味着当 λ 不是整数时众数为 ⌊λ⌋；当 λ 是整数时，众数有两个：λ 和 λ − 1。

6. Adding Independent Poisson Variables | 独立泊松变量之和

A crucial theorem for the Poisson distribution states that if X ~ Po(λ₁) and Y ~ Po(λ₂) are independent, then their sum Z = X + Y also follows a Poisson distribution with parameter λ₁ + λ₂:

泊松分布的一个重要定理指出：若 X ~ Po(λ₁) 与 Y ~ Po(λ₂) 相互独立，则它们的和 Z = X + Y 也服从参数为 λ₁ + λ₂ 的泊松分布：

If X ~ Po(λ₁) and Y ~ Po(λ₂) are independent, then X + Y ~ Po(λ₁ + λ₂)

This additivity property is extremely useful in exam problems. For instance, if the number of emails received per hour at work is Po(12) and the number received per hour at home is Po(5), and the two are independent, then the total number of emails received per hour over both locations is Po(17).

这一可加性在考试题目中极为有用。例如，若工作中每小时收到的电子邮件数服从 Po(12)，家中每小时收到的服从 Po(5)，且二者独立，那么两个地点合计每小时收到的电子邮件总数服从 Po(17)。

This property also works in reverse. If X ~ Po(λ) and we split X into two categories based on a fixed probability p, the number in the first category is Po(λp) and the number in the second category is Po(λ(1 − p)), and they are independent. This is useful when dealing with problems involving classification of events.

这一定理也可以反向使用。如果 X ~ Po(λ) 并且我们按固定概率 p 将 X 分为两类，那么第一类中的数量服从 Po(λp)，第二类中的数量服从 Po(λ(1 − p))，且二者独立。这在处理涉及事件分类的问题时很有用。

7. Poisson Approximation to the Binomial Distribution | 泊松分布对二项分布的近似

When the number of trials n is large and the probability of success p is small, the binomial distribution B(n, p) can be approximated by the Poisson distribution with λ = np. This is often called the law of rare events.

当试验次数 n 很大且成功概率 p 很小时，二项分布 B(n, p) 可以用参数 λ = np 的泊松分布来近似。这通常被称为稀有事件定律。

The approximation is generally considered acceptable when n ≥ 50 and np ≤ 5, though some textbooks give slightly different thresholds. The key point is that both the mean np and the variance np(1 − p) are close to np when p is small, because 1 − p ≈ 1.

当 n ≥ 50 且 np ≤ 5 时，近似通常被认为是可接受的，尽管不同教材给出的阈值略有不同。关键在于，当 p 很小时，二项分布的均值 np 和方差 np(1 − p) 都接近 np，因为 1 − p ≈ 1。

In Edexcel exam questions, you may be asked to use a Poisson approximation to calculate probabilities that would otherwise require lengthy binomial calculations. For example, if X ~ B(200, 0.02), then P(X ≤ 3) can be approximated by X ~ Po(4):

在 Edexcel 考试题中，你可能需要使用泊松近似来计算本来需要冗长二项式计算的概率。例如，若 X ~ B(200, 0.02)，那么 P(X ≤ 3) 可以用 X ~ Po(4) 来近似估计：

P(X ≤ 3) ≈ e⁻⁴ (1 + 4 + 4²/2! + 4³/3!) ≈ 0.433

The exact binomial calculation would require summing four individual binomial probabilities. The approximation saves time and is sufficiently accurate under the stated conditions.

精确的二项计算需要求和四个独立的二项概率。近似方法节省时间，且在给定条件下足够准确。

8. Using Tables and Technology | 查表与使用计算工具

In the Edexcel A-Level Further Mathematics exam, the Poisson distribution tables are provided for selected values of λ. These tables give cumulative probabilities P(X ≤ r) for various λ values. It is essential to be fluent in extracting both individual and cumulative probabilities from these tables.

在 Edexcel A-Level 进阶数学考试中，会提供一些选定 λ 值的泊松分布表。这些表给出不同 λ 值下的累积概率 P(X ≤ r)。熟练掌握从表中提取单个概率和累积概率至关重要。

To find P(X = r), compute P(X ≤ r) − P(X ≤ r − 1).

要求 P(X = r)，用 P(X ≤ r) − P(X ≤ r − 1) 计算。
To find P(X ≥ r), compute 1 − P(X ≤ r − 1).

要求 P(X ≥ r)，用 1 − P(X ≤ r − 1) 计算。
To find P(a ≤ X ≤ b), compute P(X ≤ b) − P(X ≤ a − 1).

要求 P(a ≤ X ≤ b)，用 P(X ≤ b) − P(X ≤ a − 1) 计算。

Many graphical calculators and statistical software also provide Poisson probabilities directly. However, in the exam you will typically use the tables provided, and clear method marks are awarded for showing the subtractions or complements you perform.

许多图形计算器和统计软件也能直接给出泊松概率。但在考试中，你通常使用提供给的表格，并且清晰的代入和减法或补事件运算可获得方法分。

9. Common Exam Applications | 常见考试应用

Exam problems involving the Poisson distribution typically fall into a few recurring categories. Familiarity with these patterns will greatly improve your speed and accuracy.

涉及泊松分布的考试题目通常可分为几类常见题型。熟悉这些模式可以显著提高你的解题速度和准确率。

Fixed interval problems: The question gives an average rate per interval and asks for the probability of a specific count. For example, cars pass a point at an average rate of 3 per minute. Find P(X = 5) in one minute. Simply substitute λ = 3 and r = 5 into the formula.

固定区间问题：题目给出每个区间的平均速率，要求某个具体计数的概率。例如，汽车以平均每分钟 3 辆的速率经过某一点，求一分钟内 X = 5 的概率。直接把 λ = 3 和 r = 5 代入公式即可。

Changed interval problems: The average rate is given for one unit, but the question asks about a different length. Scale λ proportionally. For example, if accidents occur at 2 per day, the rate for 5 days is λ = 10.

区间变换问题：题目给出某一单位的平均速率，但询问的是不同长度区间。需要按比例调整 λ。例如，若事故每天发生 2 起，5 天的速率就是 λ = 10。

Combined Poisson problems: Use the additive property. For instance, two independent Poisson processes with rates λ₁ and λ₂ combine to give a single Poisson process with rate λ₁ + λ₂.

泊松变量合并问题：利用可加性。例如，两个速率分别为 λ₁ 和 λ₂ 的独立泊松过程合并后，得到速率为 λ₁ + λ₂ 的单个泊松过程。

Approximation problems: Recognise when a binomial situation with large n and small p should be approximated by a Poisson distribution. Always state that λ = np before proceeding with Poisson calculations.

近似计算问题：识别出 n 大、p 小的二项情境可以用泊松分布近似。解题前务必先写出 λ = np。

Conditional probability problems: Use the standard formula P(A | B) = P(A ∩ B) / P(B). When the events involve Poisson counts, this often reduces to a ratio of Poisson probabilities.

条件概率问题：使用标准公式 P(A | B) = P(A ∩ B) / P(B)。当事件涉及泊松计数时，通常可化简为两个泊松概率之比。

10. Worked Example | 典型例题精解

Problem: A fair coin is flipped repeatedly. Let X represent the number of heads obtained in 300 independent flips. Use the Poisson approximation to estimate P(X ≤ 2).

题目：一枚均匀硬币被反复抛掷。设 X 表示 300 次独立抛掷中出现正面的次数。用泊松近似估计 P(X ≤ 2)。

Solution: Here, n = 300 and p = 0.5. However, p = 0.5 is not small, so the Poisson approximation is NOT appropriate for this problem. The correct approach is to use the normal approximation to the binomial, since np = 150 and n(1 − p) = 150 are both large.

解答：此处 n = 300，p = 0.5。但是 p = 0.5 并不小，所以本题不适合使用泊松近似。正确的方法是使用正态近似，因为 np = 150 和 n(1 − p) = 150 都很大。

Now consider a corrected problem: Suppose the coin is biased with P(head) = 0.01, and n = 300. Then X ~ B(300, 0.01). Since p is small and np = 3 ≤ 5, the Poisson approximation is appropriate: X ≈ Po(3).

现在考虑一个修正后的问题：假设这枚硬币有偏，P(正面) = 0.01，且 n = 300。则 X ~ B(300, 0.01)。由于 p 很小且 np = 3 ≤ 5，泊松近似适用：X ≈ Po(3)。

P(X ≤ 2) ≈ P(Y ≤ 2) where Y ~ Po(3)
= e⁻³ (1 + 3 + 9/2) = 8.5 e⁻³ ≈ 0.423

This is much simpler than computing the binomial probabilities individually. Note that the exact binomial answer is approximately 0.424, so the approximation is excellent.

这比逐个计算二项概率简单得多。精确的二项计算结果约为 0.424，近似效果非常好。

11. Common Pitfalls and Exam Tips | 常见错误与考试技巧

Students often lose marks on Poisson distribution questions due to avoidable mistakes. Here are the key pitfalls to watch out for, along with strategies to avoid them.

学生在泊松分布题目上失分，往往是因为一些可以避免的错误。以下是需要警惕的主要陷阱以及避免它们的策略。

Forgetting to change λ: When the interval changes, λ must be scaled proportionally. A rate of 4 per hour becomes λ = 8 for two hours. Many marks are lost by using the wrong λ.

忘记调整 λ：当区间改变时，λ 必须按比例缩放。每小时 4 个的速率在两小时内应为 λ = 8。许多分数因使用错误的 λ 而丢失。
Confusing P(X ≥ r) with P(X > r): For a discrete distribution, P(X > r) = P(X ≥ r + 1). Always check whether the boundary values are included or excluded.

混淆 P(X ≥ r) 与 P(X > r)：对于离散分布，P(X > r) = P(X ≥ r + 1)。务必检查边界值是否包含在内。
Incorrect table reading: When using cumulative tables, remember that P(X = r) is obtained by a subtraction. Never read the table value directly as P(X = r).

表格读取错误：使用累积概率表时，记住 P(X = r) 需要做减法。切勿将表值直接当作 P(X = r)。
Misusing the binomial approximation: Only use the Poisson approximation when n is large and p is small (typically n ≥ 50 and np ≤ 5). If p is moderate or large, consider a normal approximation instead, provided np and n(1−p) are both large.

误用二项近似：只有在 n 大且 p 小（通常 n ≥ 50 且 np ≤ 5）时才使用泊松近似。如果 p 适中或较大，考虑使用正态近似，前提是 np 和 n(1−p) 都较大。
Forgetting the assumptions: In questions that ask whether a Poisson model is suitable, explicitly test each assumption: independence, constant rate, and rarity of simultaneous events.

忘记验证假设：在问泊松模型是否合适的题目中，需逐一检验每个假设：独立性、恒定性以及同时事件极少的条件。

12. Summary and Final Advice | 总结与建议

The Poisson distribution is a cornerstone of A-Level Further Mathematics statistics. Its defining feature is that the mean equals the variance, both being λ. The probability mass function P(X = r) = e⁻λ λʳ / r! allows direct computation of probabilities for any non-negative integer r.

泊松分布是 A-Level 进阶数学统计部分的基石。其定义性特征在于均值等于方差，均为 λ。概率质量函数 P(X = r) = e⁻λ λʳ / r! 允许直接计算任何非负整数 r 的概率。

Remember the key assumptions: independent events, constant average rate, and no simultaneous occurrences. These are not mere formalities — they determine whether the Poisson model is legitimate for a given real-world scenario.

请记住关键假设：事件独立、平均速率恒定、不发生同时事件。这些不仅仅是形式——它们决定了泊松模型在特定现实场景中是否合理。

In the exam, always check the value of λ carefully, pay attention to whether cumulative or individual probabilities are required, and use the additive property when combining independent Poisson variables. With consistent practice, Poisson distribution problems become some of the most straightforward marks available in the Further Mathematics paper.

在考试中，务必仔细检查 λ 的值，注意题目要求的是累积概率还是单个概率，并在合并独立泊松变量时使用可加性。通过持续练习，泊松分布题会成为进阶数学试卷中最好拿分的题目之一。

Published by TutorHao | Further Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Poisson Distribution: Basic Concepts and Characteristics | 泊松分布的基本概念与特征