📚 Mean and Variance of a Poisson Distribution | 泊松分布的均值与方差
The Poisson distribution is a cornerstone of probability theory, especially in A-Level Edexcel Mathematics. It elegantly models the number of random, independent events occurring in a fixed interval of time or space. A deep understanding of its mean and variance is essential, not only for solving exam problems but also for grasping why this distribution behaves as it does. Both the mean and variance equal the parameter λ, a property that sets the Poisson distribution apart from many others.
泊松分布是概率论的基石，特别是在 A-Level 爱德思数学中。它优雅地模拟了在固定时间或空间间隔内随机、独立事件发生的次数。深入理解其均值和方差至关重要，这不仅是解考试题目的需要，也是理解该分布行为的关键。均值和方差都等于参数 λ，这个性质让泊松分布与众不同。
1. The Poisson Distribution: A Quick Recap | 泊松分布快速回顾
discrete random variable following a Poisson distribution with parameter λ > 0. We write X ~ Po(λ). The probability mass function (PMF) is P(X = x) = e⁻λ λˣ / x! er of calls arriving at a call centre in an hour, or the number of defects in a meter of fabric.
设 X 是一个服从参数为 λ (>0) 的泊松分布的离散随机变量，记为 X ~ Po(λ)。其概率质量函数为 P(X = x) = e⁻λ λˣ / x!，其中 x = 0, 1, 2, ... 。事件独立发生且平均速率为常数 λ。该分布常用于模拟稀有事件，比如一小时内打进呼叫中心的电话数，或者一米织物上的瑕疵数。
2. The Parameter λ: The Rate of Occurrence | 参数 λ：发生速率
λ represents both the expected number of occurrences in the given interval and the variance of the distribution. It is the only parameter, so once you know the average rate, the entire shape of the distribution is determined. In real-world scenarios, λ is often estimated from sample data as the sample mean number of events per interval.
λ 既代表在给定区间内事件发生的期望次数，也代表该分布的方差。它是唯一的参数，因此一旦知道平均速率，整个分布的形状就确定了。在实际情况中，λ 通常由样本数据估计得出，即每个区间内事件发生的样本均值。
3. Deriving the Mean: E(X) = λ | 推导均值：E(X) = λ
x values. For the Poisson distribution, we start with E(X) = Σ (from x=0 to ∞) x · (e⁻λ λˣ / x!). Note that the term for x = 0 is zero, so we can begin the sum at x=1.
根据定义，离散随机变量的期望值为 E(X) = Σ x · P(X = x)，对所有可能的 x 值求和。对于泊松分布，我们从 E(X) = Σ (x=0 到 ∞) x · (e⁻λ λˣ / x!) 开始。注意 x = 0 的项为零，因此可以从 x=1 开始求和。
Rewrite the expression: E(X) = e⁻λ Σ (x=1 to ∞) x · (λˣ / x!). Since x / x! = 1 / (x-1)!, we have E(X) = e⁻λ Σ (x=1 to ∞) λˣ / (x-1)! hift the index by letting k = x-1, so when x=1, k=0. The sum becomes Σ (k=0 to ∞) λᵏ⁺¹ / k! = λ Σ (k=0 to ∞) λᵏ / k!.
重写表达式：E(X) = e⁻λ Σ (x=1 到 ∞) x · (λˣ / x!)。因为 x / x! = 1 / (x-1)!，所以 E(X) = e⁻λ Σ (x=1 到 ∞) λˣ / (x-1)!。现在进行指标平移，令 k = x-1，那么当 x=1 时 k=0。求和变为 Σ (k=0 到 ∞) λᵏ⁺¹ / k! = λ Σ (k=0 到 ∞) λᵏ / k!。
The remaining sum is the Maclaurin series for e²: Σ (k=0 to ∞) λᵏ / k! = λ. This elegant derivation shows why the mean of a Poisson random variable is exactly λ.
剩余的求和是 e² 的麦克劳林级数：Σ (k=0 到 ∞) λᵏ / k! = e²。因此，E(X) = e⁻λ · λ e² = λ。这个优雅的推导过程展示了为什么泊松随机变量的均值恰好为 λ。
4. Step-by-Step Summation for E(X) | 逐项求和计算 E(X) 的细节
To avoid confusion, let's lay out the key algebraic steps clearly in a centred display. The crucial simplification is that x/x! simplifies to 1/(x-1)!, allowing the sum to be re-indexed. The series expansion of e² then collapses the infinite sum to a finite expression.
为避免混淆，我们用一个居中的展示把关键代数步骤清晰地列出来。关键的简化是 x/x! 可以简化为 1/(x-1)!，从而允许重新索引。e² 的级数展开将无穷和压缩为一个有限表达式。
E(X) = Σ (x=1 to ∞) x e⁻λ λˣ / x! = e⁻λ Σ (x=1 to ∞) λˣ / (x-1)! = e⁻λ λ Σ (k=0 to ∞) λᵏ / k! = e⁻λ λ e² = λ
This technique of shifting the summation index is a standard tool in A-Level exam proofs. Practice manipulating factorial terms to build confidence.
这种平移求和索引的技巧是 A-Level 考试证明中的标准工具。练习操作阶乘项以建立信心。
5. Deriving the Variance: Var(X) = λ | 推导方差：Var(X) = λ
. Since we already know E(X) = λ, we now need E(X²). Compute E(X²) = Σ (x=0 to ∞) x² e⁻λ λˣ / x! mmon trick is to first find E[X(X-1)], because x² = x(x-1) + x, making the algebra neater.
方差定义为 Var(X) = E(X²) - [E(X)]²。既然我们已知 E(X) = λ，现在需要求 E(X²)。计算 E(X²) = Σ (x=0 到 ∞) x² e⁻λ λˣ / x!。一个常见技巧是先求 E[X(X-1)]，因为 x² = x(x-1) + x，这样代数运算更整洁。
Thus, E[X(X-1)] = E(X²) - E(X), so E(X²) = E[X(X-1)] + λ. Let's find E[X(X-1)] = Σ (x=0 to ∞) x(x-1) e⁻λ λˣ / x!. Again, terms for x=0 and x=1 vanish because x(x-1)=0, so we start at x=2.
因此，E[X(X-1)] = E(X²) - E(X)，所以 E(X²) = E[X(X-1)] + λ。现在求 E[X(X-1)] = Σ (x=0 到 ∞) x(x-1) e⁻λ λˣ / x!。同样，x=0 和 x=1 的项为零，因为 x(x-1)=0，所以我们从 x=2 开始。
Rewrite: E[X(X-1)] = e⁻λ Σ (x=2 to ∞) x(x-1) λˣ / x! = e⁻λ Σ (x=2 to ∞) λˣ / (x-2)!. Set k = x-2, giving E[X(X-1)] = e⁻λ Σ (k=0 to ∞) λᵏ⁺² / k! = e⁻λ λ² Σ (k=0 to ∞) λᵏ / k! = e⁻λ λ² e² = λ².
重写：E[X(X-1)] = e⁻λ Σ (x=2 到 ∞) x(x-1) λˣ / x! = e⁻λ Σ (x=2 到 ∞) λˣ / (x-2)!。令 k = x-2，得到 E[X(X-1)] = e⁻λ Σ (k=0 到 ∞) λᵏ⁺² / k! = e⁻λ λ² Σ (k=0 到 ∞) λᵏ / k! = e⁻λ λ² e² = λ²。
6. Completing the Variance Calculation | 完成方差计算
+ λ) - λ² = λ. This proves that the variance of a Poisson distribution equals its mean, λ.
现在我们得到 E[X(X-1)] = λ²。因此，E(X²) = λ² + λ。代入方差公式得到 Var(X) = (λ² + λ) - λ² = λ。这证明了泊松分布的方差等于其均值 λ。
Var(X) = E(X²) - [E(X)]² = (λ² + λ) - λ² = λ
he result E[X(X-1)] = λ² can save time, but you must be ready to reproduce the derivation.
阶乘矩的方法极其高效，经常在爱德思试卷中出现。记住结果 E[X(X-1)] = λ² 可以节省时间，但你必须准备好重现推导过程。
7. The Elegant Result: Mean = Variance | 优雅的结果：均值等于方差
One of the identifying features of a Poisson distribution is that its mean and variance are identical. This property is not shared by the binomial or geometric distributions (except in limiting cases). In real-world data analysis, if the sample mean and sample variance are roughly equal, the Poisson model is often appropriate. However, if the variance is significantly larger than the mean (overdispersion), a different distribution may be needed.
泊松分布的一个标志性特征是其均值与方差相等。二项分布或几何分布不具备这个性质（除了极限情形）。在实际数据分析中，如果样本均值与样本方差大致相等，泊松模型通常是合适的。但如果方差明显大于均值（过度离散），则可能需要其他分布。
8. Worked Example: Applying Mean and Variance | 计算示例：应用均值与方差
iger counter records radioactive decays at an average rate of 3.2 decays per second. Let X ~ e second. Immediately, we can state E(X) = 3.2 and Var(X) = 3.2. The standard deviation is √3.2 ≈ 1.789.
假设一个盖革计数器以每秒平均 3.2 次衰变的速率记录放射性衰变。令 X ~ Po(3.2) 表示一秒内的衰变次数。我们可以直接写出 E(X) = 3.2 且 Var(X) = 3.2。标准差为 √3.2 ≈ 1.789。
f we want the probability of exactly 2 decays, we use the PMF: P(X = 2) = e⁻³·² (3.2)² / 2! us what to expect on average, while the variance indicates the spread around that average.
现在，如果我们想求恰好发生 2 次衰变的概率，使用概率质量函数：P(X = 2) = e⁻³·² (3.2)² / 2! = (0.04076 × 10.24) / 2 ≈ 0.2087。均值告诉我们平均预期是什么，而方差指示围绕该均值的离散程度。
Po(32), with E(Y) = Var(Y) = 32. These scaling properties arise directly from the mean and variance formulas.
此外，在抽样情境中，如果我们观察了 10 个独立的一秒区间，10 秒内的总计数 Y 将服从 Po(10×3.2) = Po(32)，其 E(Y) = Var(Y) = 32。这些缩放性质直接源自均值和方差公式。
9. Common Pitfalls and Misconceptions | 常见错误与误解
-something). Another frequent error is forgetting to simplify factorials correctly when deriving E(X) or Var(X), especially missing that x/x! = 1/(x-1)!. Always start summing from the first non-zero term.
学生经常把方差公式与二项分布的 npq 混淆。记住：泊松分布的方差就是 λ，而不是 λ(1-某值)。另一个常见错误是在推导 E(X) 或 Var(X) 时没有正确地简化阶乘，尤其是忘记了 x/x! = 1/(x-1)!。务必从第一个非零项开始求和。
The equality of mean and variance does not imply equality of mean and standard deviation.
另外，有些学习者错误地认为标准差是 λ，而实际上是 √λ。均值与方差相等并不意味着均值与标准偏差相等。
mial distribution. In such cases, the mean and variance of the approximating Poisson are both np, while the true binomial variance is np(1-p). The approximation works well when n is large and p is small, making np(1-p) close to np.
最后，注意近似时的处理：当 n 很大且 p 很小时，泊松分布 (λ=np) 可以近似二项分布。此时，近似泊松分布的均值和方差都是 np，而真实二项分布的方差是 np(1-p)。当 n 很大且 p 很小时，近似效果很好，因为 np(1-p) 接近 np。
10. Exam Tips and Key Points | 考试技巧与重点
he summation index shift and the series expansion of e². Examiners look for correct use of 'let k = x-1' or similar steps. Always state the domains clearly.
在爱德思 A-Level 考试中，你可能会被要求从基本原理证明 E(X) = λ 或 Var(X) = λ。确保你能够清晰地展示求和索引平移和 e² 的级数展开。考官会注意你是否正确使用了“令 k = x-1”或类似步骤。始终清晰地标明定义域。
When tackling problems involving the difference or sum of independent Poisson variables, remember that the mean and variance add up: if X ~ Po(λ₁) and Y ~ Po(λ₂) are independent, then X+Y ~ Po(λ₁+λ₂). Its mean and variance are both λ₁+λ₂. This directly follows from the mean and variance properties.
在处理涉及独立泊松变量之和与差的问题时，记住均值与方差可加：如果 X ~ Po(λ₁) 和 Y ~ Po(λ₂) 独立，则 X+Y ~ Po(λ₁+λ₂)。其均值和方差均为 λ₁+λ₂。这直接源于均值和方差的性质。
For hypothesis testing with a Poisson mean, you may need to find critical regions using λ under H₀. Because variance equals mean, the test statistic's distribution is fully specified under the null hypothesis, which simplifies calculations.
在对泊松均值进行假设检验时，你可能需要使用 H₀ 下的 λ 寻找临界区域。因为方差等于均值，在原假设下检验统计量的分布是完全确定的，这简化了计算。
11. Summary | 总结
The Poisson distribution with parameter λ is unique in that its mean and variance are both equal to λ. Derivations rely on manipulating the infinite sum using factorial simplifications and the series for e². Mastery of these derivations not only secures marks in proof-based exam questions but also deepens your understanding of discrete probability models. Keep revising the step-by-step index shift and practice with different λ values to build fluency.
参数为 λ 的泊松分布独特之处在于其均值和方差都等于 λ。推导过程依赖于使用阶乘简化和 e² 级数对无穷和进行操作。掌握这些推导不仅能在证明类考题中稳拿分数，还能加深你对离散概率模型的理解。反复练习逐步索引平移，并使用不同的 λ 值进行练习，以提升熟练度。
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