📚 The Inverse Normal Distribution Function | 逆正态分布函数
In A-Level Mathematics, the normal distribution is a cornerstone of statistical modelling. Its cumulative distribution function (CDF) allows us to find the probability that a randomly selected value falls below a given point. However, many real-world problems begin with a known probability and ask us to find the corresponding value — this is where the inverse normal distribution function becomes essential. Also known as the quantile function, it reverses the usual calculation: given a left‑tail probability, it returns the z‑score or raw data value that cuts off that area under the bell curve. Mastering this tool is key for tackling Edexcel exam questions on normal distribution, from finding percentiles to setting warranty limits.
在A-Level数学中，正态分布是统计建模的基石。其累积分布函数（CDF）让我们能够求出随机变量取值落在某一点以下的概率。然而，许多现实问题是从已知的概率出发，要求我们找出对应的数值——这便是逆正态分布函数发挥作用的地方。它也被称为分位数函数，把常规计算颠倒过来：给定左尾概率，它会返回在钟形曲线下截取该面积的z分数或原始数据值。掌握这个工具是解决Edexcel正态分布考试题的关键，无论是求百分位数还是设定保修期限。
1. What Is the Inverse Normal Distribution Function? | 什么是逆正态分布函数？
Φ(z). If Φ(z) = P(Z ≤ z) = p, then the inverse function, often written Φ⁻¹(p), gives the z‑ ts left under the standard normal curve equals p. In other words, it answers the question: 'What z‑score corresponds to a cumulative probability of p?' For any normal distribution X ~ N(μ, σ²), the inverse normal is applied after standardising back from a z‑score to the original units using X = μ + σz.
逆正态分布函数是标准正态累积分布函数Φ(z)的反函数。如果Φ(z) = P(Z ≤ z) = p，那么逆函数（常写作Φ⁻¹(p)）会给出使得标准正态曲线下左侧面积等于p的z值。换句话说，它回答了这个问题：“累积概率p对应的z分数是多少？”对于任意正态分布X ~ N(μ, σ²)，逆正态用于从z分数通过X = μ + σz还原为原始单位的数值。
2. The Standard Normal Distribution and Its Inverse | 标准正态分布及其逆函数
rea to the left of 1.96 under the standard normal curve is exactly 0.975. Many statistical tables provide a limited set of inverse values, but modern calculators compute Φ⁻¹(p) for any p between 0 and 1.
标准正态分布的均值为0，标准差为1。其累积概率有大量的表格可查，但表格只能单向使用：从z查到概率。逆函数Φ⁻¹(p)使我们能反过来从概率得到z值。例如，Φ⁻¹(0.975) = 1.96，因为在标准正态曲线下，位于1.96左侧的面积恰好为0.975。许多统计表提供一套有限的逆值，但现代计算器可以计算0到1之间任意p对应的Φ⁻¹(p)。
3. Notation and Key Properties | 符号与关键性质
m the bell curve being symmetric about zero. In Edexcel exams, you will often use this property to find lower‑tail critical values quickly.
符号Φ⁻¹(p)并不表示1/Φ(z)；它是一个反函数。它的定义域为(0, 1)，值域为全体实数。一个值得记住的性质是对称性：Φ⁻¹(1 − p) = −Φ⁻¹(p)。例如，Φ⁻¹(0.025) = −1.96，这与我们熟悉的Φ⁻¹(0.975) = 1.96相呼应。这种对称性源于钟形曲线关于零点对称。在Edexcel考试中，你会经常利用这个性质快速求出下尾临界值。
4. Using a Calculator: InvNorm on Casio and TI | 计算器操作：Casio与TI的InvNorm功能
ction for the inverse normal. Typically accessed via the distribution menu, it is labelled 'InvNorm' or 'Inverse Normal'. The user inputs the area (left‑tail probability), the mean, and the standard deviation. If only the z‑ore is desired, set mean = 0 and standard deviation = 1. The calculator then returns the x‑value or z‑ monstrate proficiency with this function, often writing down the input parameters clearly.
现代科学计算器，如Casio fx-991EX或TI-84 Plus，都有专门的逆正态函数。通常通过分布菜单访问，标签为“InvNorm”或“逆正态”。用户输入面积（左尾概率）、均值和标准差。如果只需要z分数，将均值设为0、标准差设为1。计算器随即返回x值或z分数。在Edexcel考试中，要求你熟练使用此功能，通常需要清晰地写出输入的参数。
5. From Probability to z‑score: A Detailed Look | 从概率到z分数：深入解析
verse function, k = Φ⁻¹(0.8). Using a calculator, we obtain k ≈ 0.8416. This means that 80% body of the normal table for 0.8 (or the closest value), then read off the corresponding z‑value, but the calculator gives far greater precision.
假设我们已知标准正态随机变量小于某个未知值k的概率为0.8。我们写成P(Z < k) = 0.8。应用逆函数，得k = Φ⁻¹(0.8)。使用计算器，求得k ≈ 0.8416。这意味着标准正态曲线下80%的面积位于0.8416的左侧。在查表时，你需要在正态表格的内部查找0.8（或最接近的值），然后读出相应的z值，但计算器能提供高得多的精度。
6. Percentiles and the Inverse Normal | 百分位数与逆正态
n of the inverse normal. The 90th percentile of a distribution is the value below which 90% . For a standard normal distribution, the 90th percentile is Φ⁻¹(0.90) ≈ 1.2816. For a non‑ say IQ scores with μ = 100, σ = 15, the 90th percentile is found by first obtaining the z‑score for 0.90 and then transforming: x = 100 + 15 × 1.2816 ≈ 119.2. This two‑step process — find z then un‑standardise — is a core skill tested in Edexcel exams.
百分位数是逆正态的一个自然应用。一个分布的第90百分位数是指有90%的数据落在其下方的那个数值。对于标准正态分布，第90百分位数为Φ⁻¹(0.90) ≈ 1.2816。对于非标准正态分布，比如IQ分数（μ = 100，σ = 15），第90百分位数的求法是：先得到对应0.90的z分数，再进行变换：x = 100 + 15 × 1.2816 ≈ 119.2。这个两步过程——先求z再反标准化——是Edexcel考试中考查的核心技能。
7. The General Inverse Normal: Un‑standardising | 一般逆正态：反标准化
For any normal distribution X ~ z = (x − μ)/σ. When solving problems, always check whether the given probability is a left‑tail or right‑tail probability. If a question gives a right‑tail probability (e.g., top 5%), subtract it from 1 to obtain the left‑tail area: p = 1 − 0.05 = 0.95, and then apply Φ⁻¹(0.95).
对于任意正态分布X ~ N(μ, σ²)，计算器上的逆正态函数可以直接接受μ和σ，并返回原始分数x。其关系式为 x = μ + σ · Φ⁻¹(p)。该公式源自标准化 z = (x − μ)/σ。在解题时，务必检查给定的概率是左尾概率还是右尾概率。如果题目给出右尾概率（如前5%），则需要用1减去它来得到左尾面积：p = 1 − 0.05 = 0.95，然后再应用Φ⁻¹(0.95)。
8. Worked Example: Ball Bearings and Quality Control | 实例精讲：滚珠轴承与质量控制
A machine produces ball bearings whose diameters are normally distributed with mean 5.00 mm and standard deviation 0.02 mm. Bearings are rejected if their diameter is in the smallest 2% of the distribution. Find the minimum acceptable diameter.
一台机器生产的滚珠轴承直径服从正态分布，均值为5.00 mm，标准差为0.02 mm。如果轴承的直径落在分布中最小的2%范围内，则被拒收。求可接受的最小直径。
The smallest 2% corresponds to a left‑tail probability of 0.02. We need the 2nd percentile. Using the calculator's InvNorm with area = 0.02, μ = 0, σ = 1, we get z = Φ⁻¹(0.02) ≈ −2.0537. Then un‑ are rejected. Always remember to round appropriately in context; here, perhaps to 4.96 mm.
最小的2%对应左尾概率0.02。我们需要求第2百分位数。使用计算器的InvNorm，面积=0.02，μ=0，σ=1，得到z = Φ⁻¹(0.02) ≈ −2.0537。然后反标准化：x = 5.00 + 0.02 × (−2.0537) = 5.00 − 0.04107 = 4.95893 mm。因此，直径小于约4.96 mm的轴承会被拒收。在上下文中要记得适当舍入；这里或许可以舍入为4.96 mm。
9. Common Pitfalls and How to Avoid Them | 常见陷阱及对策
One of the most frequent errors is confusing the left‑tail probability with the right‑tail one. If the question asks for the value such that 10% of items exceed it, the left‑tail area is 0.90, not 0.10. Another pitfall is misapplication of the symmetry property: Φ⁻¹(0.05) is not 1.6449 but −1.6449. Also, students sometimes forget to set the calculator to the correct mean and standard deviation, leading to raw scores when z‑scores are intended, or vice versa. Always write down the parameters you enter into InvNorm as part of your working.
最常见的错误之一是将左尾概率与右尾概率混淆。如果题目要求找出有10%的观测值超过它的那个数值，那么左尾面积就是0.90，而不是0.10。另一个陷阱是错误使用对称性质：Φ⁻¹(0.05)不是1.6449，而是−1.6449。此外，学生有时会忘记在计算器上设置正确的均值和标准差，导致本想求z分数却得到了原始分数，或者反过来。务必把你输入InvNorm的参数作为解题步骤的一部分写下来。
10. Handling Two‑Tailed Problems with the Inverse Normal | 用逆正态处理双侧问题
In hypothesis testing or confidence intervals, two‑ common. For a significance level α = 0.05, each tail contains α/2 = 0.025. The critical z‑ sing a calculator, you can find either value and simply change the sign. Recognising this pattern saves time and reduces error in exams.
在假设检验或置信区间中，双侧情形十分常见。对于显著性水平α = 0.05，每个尾部包含α/2 = 0.025。临界z值为±Φ⁻¹(0.975) = ±1.96。这展示了对性：下临界z为Φ⁻¹(0.025) = −1.96，上临界z为Φ⁻¹(0.975) = 1.96。使用计算器时，你可以求出其中一个值，然后直接改变符号。认识这种模式可以节省考试时间并减少错误。
11. Exam Tips for Edexcel A‑Level Mathematics | Edexcel A‑Level数学考试贴士
xcel papers, especially in the Statistics part of the course. Look for questions involving 'find the value such that...', 'the top x%', 'the quartiles', or 'warranty limits' the relevant area to confirm the probability direction. Show clear notation: P(X < k) = p ⇒ given. Finally, interpret your answer in the context of the problem with a short sentence.
逆正态函数经常出现在Edexcel试卷中，尤其是课程的统计部分。留意那些“求使得……的值”、“最高的x%”、“四分位数”或“保修期限”的题目。始终快速画一条正态曲线并给相应区域涂上阴影，以确认概率的方向。展示清晰的符号：对于标准正态，P(X < k) = p ⇒ k = Φ⁻¹(p)；当给出参数时，k = μ + σΦ⁻¹(p)。最后，用简短的句子在题目情境中解释你的答案。
12. Summary: The Inverse Normal as a Reverse Lookup | 总结：逆正态作为反向查找工具
The inverse normal distribution function is simply the mechanism for turning a known cumulative probability into the corresponding quantile of the normal distribution. Whether you are working with the standard normal Z or a general normal X, the procedure remains consistent: identify the correct left‑tail probability, apply Φ⁻¹ (or calculator InvNorm) to find the z‑score, and un‑standardise if needed. A sound grasp of this function not only boosts your performance on normal distribution questions but also underpins later topics such as confidence intervals and hypothesis tests.
逆正态分布函数不过是将已知的累积概率转换为正态分布相应分位数的机制。无论你处理的是标准正态Z还是一般正态X，步骤始终一致：确定正确的左尾概率，应用Φ⁻¹（或计算器InvNorm）求出z分数，若需要再反标准化。透彻掌握这个函数不仅能提升你在正态分布题上的表现，也为后续的置信区间和假设检验等专题打下基础。
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