📚 Modelling with the Poisson Distribution | 泊松分布建模
The Poisson distribution is a discrete probability distribution that expresses the probability of a given number of events occurring in a fixed interval of time or space, provided these events happen with a known constant mean rate and independently of the time since the last event. In A-level Mathematics, modelling with the Poisson distribution allows us to analyse real-world situations such as the number of calls arriving at a call centre per hour, the number of flaws in a length of wire, or the number of accidents at a junction per week. Understanding when and how to apply the Poisson model is essential for success in Edexcel Statistics, and it forms the foundation for further work in hypothesis testing and approximations.
泊松分布是一种离散概率分布，用于表示在固定的时间或空间区间内，事件发生的次数，前提是这些事件以已知的恒定平均速率发生，并且与上次事件以来的时间无关。在A-Level数学中，利用泊松分布建模使我们能够分析现实世界的情况，比如每小时呼叫中心接到的电话数量、一段电线上的瑕疵数量，或者每周某路口的事故数量。理解何时以及如何应用泊松模型对于在Edexcel统计学中取得成功至关重要，并且它为假设检验和近似计算等进一步学习奠定了基础。
1. Introduction to the Poisson Distribution | 泊松分布简介
The Poisson distribution is named after the French mathematician Simeon Denis Poisson. It models the number of times an event occurs in a given interval, where the events are rare and random. The distribution has a single parameter, λ (lambda), which represents the mean number of occurrences in that interval. In the Edexcel specification, you will encounter the notation X ~ Po(λ) to indicate that the random variable X follows a Poisson distribution with mean λ.
泊松分布以法国数学家西莫恩·德尼·泊松的名字命名。它用于模拟在给定区间内事件发生的次数，其中事件是稀有且随机的。该分布只有一个参数λ（lambda），表示在该区间内事件发生的平均次数。在Edexcel考试大纲中，你会遇到记号X ~ Po(λ)，表示随机变量X服从均值为λ的泊松分布。
2. Conditions for a Poisson Model | 泊松模型的条件
For a situation to be modelled by a Poisson distribution, three key conditions must be satisfied. First, events must occur singly in time or space – two events cannot happen at exactly the same instant. Second, events must occur at a constant average rate, meaning the value of λ remains uniform throughout the interval. Third, events must be independent, so the occurrence of one event does not affect the probability of another occurring. If these conditions hold, the Poisson model is appropriate.
要用泊松分布对某种情况进行建模，必须满足三个关键条件。第一，事件必须在时间或空间上单独发生——两个事件不能恰好在同一瞬间发生。第二，事件必须以恒定的平均速率发生，这意味着λ的值在整个区间内保持一致。第三，事件必须相互独立，因此一个事件的发生不会影响另一个事件发生的概率。如果这些条件成立，那么泊松模型就是适用的。
3. The Poisson Probability Formula | 泊松概率公式
The probability of observing exactly x events in a given interval is given by the formula:
P(X = x) = (e⁻λ λˣ) / x! for x = 0, 1, 2, ...
where e is the base of natural logarithms (approximately 2.71828), λ is the mean number of occurrences, and x! denotes x factorial. This formula is provided in the Edexcel formula booklet, but you must be able to substitute values correctly and use your calculator to evaluate probabilities, including cumulative probabilities.
在给定区间内恰好观测到x次事件的概率由下式给出：
P(X = x) = (e⁻λ λˣ) / x! 对于 x = 0, 1, 2, ...
其中e是自然对数的底数（约等于2.71828），λ是事件发生的平均次数，x!表示x的阶乘。该公式在Edexcel的公式手册中给出，但你必须能够正确代入数值，并利用计算器计算概率，包括累积概率。
4. Mean and Variance of a Poisson Distribution | 泊松分布的均值和方差
special property of the Poisson distribution is that its mean and variance are equal. If X ~ ughly equal, a Poisson model may be suitable. The standard deviation is therefore √λ. In modelling, this relationship helps us estimate λ from data by using the sample mean.
泊松分布的一个特殊性质是其均值与方差相等。如果X ~ Po(λ)，那么期望值E(X) = λ，方差Var(X) = λ。当分析数据时，这种相等性提供了一个快速检验方法：如果样本均值和方差大致相等，那么泊松模型可能是合适的。因此标准差为√λ。在建模中，这种关系有助于我们通过样本均值来估计λ。
5. Modelling Events in Time | 时间维度的事件建模
an average of 4 calls per hour. We can define X as the number of calls in one hour, and X ~ probability of receiving exactly 2 calls in an hour, we compute P(X = 2) = (e⁻⁴ × 4²) / 2! ≈ 0.1465. Such models help businesses plan staffing levels and resources.
一个常见的应用是对固定时间段内发生的事件数量进行建模。例如，一个呼叫中心平均每小时接到4个电话。我们可以定义X为一小时内的电话数量，且X ~ Po(4)。为了求出一小时内恰好接到2个电话的概率，我们计算P(X = 2) = (e⁻⁴ × 4²) / 2! ≈ 0.1465。这样的模型有助于企业规划人员编制和资源配置。
6. Modelling Events in Space | 空间维度的事件建模
The Poisson distribution is equally useful for events distributed over space, such as flaws on a production line or organisms in a volume of liquid. Suppose a fabric has an average of 0.5 defects per square metre. Then the number of defects in 10 square metres, Y, follows Po(5). The model treats area just like time – the interval is simply spatial. The conditions of randomness, uniformity and independence must still hold.
泊松分布同样适用于分布在空间上的事件，例如生产线上的瑕疵或一定体积液体中的微生物数量。假设一种织物平均每平方米有0.5个疵点。那么10平方米内的疵点数量Y服从Po(5)。该模型将面积视为和时间一样——区间仅仅是空间上的。随机性、均匀性和独立性条件仍然必须满足。
7. Changing the Interval | 更改区间
proportional to the length. For instance, if calls arrive at a rate of 6 per hour, then for a 30‑minute period λ = 3, and for a 2‑hour period λ = 12. Mathematically, if X ~ scaling property makes the Poisson model very flexible for different observation windows.
如果某个区间内的事件数量服从泊松分布，你可以轻松地针对不同的区间长度调整λ。新参数与区间长度成正比。例如，如果电话以每小时6个的速率到达，那么对于30分钟的时段λ = 3，对于2小时的时段λ = 12。在数学上，如果在一个时间单位内X ~ Po(λ)，那么对于t个时间单位，分布为Po(λt)。这种缩放特性使得泊松模型对于不同的观测窗口非常灵活。
8. Sum of Independent Poisson Variables | 独立泊松变量之和
Another important modelling property is that the sum of two or more independent Poisson random variables is also Poisson distributed, with its mean equal to the sum of the individual means. If X ~ Po(λ₁) and Y ~ Po(λ₂) are independent, then X + Y ~ Po(λ₁ + λ₂). This result enables us to combine independent Poisson processes, such as the total number of arrivals from two separate sources, into a single Poisson model, greatly simplifying analysis.
另一个重要的建模性质是，两个或多个独立的泊松随机变量之和也服从泊松分布，其均值等于各个均值之和。如果X ~ Po(λ₁) 和 Y ~ Po(λ₂) 相互独立，那么X + Y ~ Po(λ₁ + λ₂)。这一结论使我们能够将独立的泊松过程（例如来自两个独立来源的到达总数）合并为一个单一的泊松模型，从而极大地简化分析。
9. Poisson Approximation to the Binomial | 泊松分布对二项分布的近似
tion is used when n ≥ 50 and np ≤ 5. For example, if a factory produces 1000 items and the probability of a defective item is 0.003, the number of defectives X ~ B(1000, 0.003) can be approximated by X ~ error is minimal. This is a key application of Poisson modelling in statistical inference.
当n很大而p很小时，二项分布B(n, p)可以用均值为λ = np的泊松分布来近似。通常，当n ≥ 50且np ≤ 5时使用这种近似。例如，如果一个工厂生产1000件产品，且每件产品为次品的概率为0.003，那么次品数量X ~ B(1000, 0.003)可以近似为X ~ Po(3)。这样计算会变得简单得多，而且误差极小。这是泊松建模在统计推断中的一个重要应用。
10. Hypothesis Testing with Poisson | 泊松分布的假设检验
Modelling is often followed by hypothesis testing to determine whether observed data are consistent with a Poisson model. In Edexcel questions, you may be given a claimed value of λ and asked to test this at a significance level. Using cumulative Poisson probabilities, you find the critical region for the test statistic. For a two‑tailed test, you split the significance level between the lower and upper tails. If the observed count falls inside the critical region, the null hypothesis is rejected, suggesting the Poisson model is not appropriate.
建模之后往往要进行假设检验，以判断观测数据是否与泊松模型一致。在Edexcel的考题中，你可能会得到一个声称的λ值，并被要求在某个显著性水平下进行检验。利用累积泊松概率，你可以找到检验统计量的临界域。对于双尾检验，你将显著性水平分配到下尾和上尾。如果观测计数落入临界域内，则拒绝原假设，表明该泊松模型并不合适。
11. Choosing an Appropriate Model | 选择合适模型
Choosing to model a situation with the Poisson distribution requires careful judgement. You should check the three conditions explicitly and consider whether the data show over‑dispersion (variance notably larger than the mean), which would violate the Poisson assumption. Additionally, if events are not independent or occur in clusters, a Poisson model is unsuitable. In exam questions, you will often be asked to state assumptions and comment on the suitability of the Poisson model in context.
选择用泊松分布对某种情形进行建模需要仔细判断。你应该明确检查三个条件，并考虑数据是否显示出过度离散（方差明显大于均值），这将违反泊松假设。此外，如果事件不独立或成簇出现，泊松模型就不适用。在考试题目中，你经常会被要求陈述假设，并结合背景评论泊松模型的适用性。
12. Real-World Applications and Limitations | 实际应用与局限
The Poisson distribution is widely used in areas such as insurance (number of claims per year), traffic engineering (cars passing a point per minute), biology (mutations in a DNA strand), and queueing theory (customers arriving at a till). However, its limitation lies in the assumption of a constant rate; if the rate changes over time, a non‑homogeneous Poisson process is needed. At A‑level, you work with the basic homogeneous model, which provides a powerful introduction to stochastic modelling and lays the groundwork for further study.
泊松分布广泛应用于保险（每年索赔次数）、交通工程（每分钟通过某点的车辆数）、生物学（DNA链中的突变）以及排队论（到达收银台的顾客数）等领域。然而，其局限性在于恒定速率的假设；如果速率随时间变化，则需要非齐次泊松过程。在A-Level阶段，你主要学习基本的齐次模型，它为随机建模提供了有力的入门，并为进一步的学习奠定了基础。
Published by TutorHao | 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