📚 Exercise 18F: Normal Distribution and Statistical Inference | 练习18F：正态分布与统计推断
In IB Mathematics, Exercise 18F typically challenges you to apply the properties of the normal distribution to real-world data, calculate probabilities using a standardized approach, and make inferential statements about populations. This exercise consolidates your ability to work with continuous random variables, use the inverse normal function, and understand the conditions under which the normal model is appropriate. Mastering 18F means you can confidently move between raw data, z‑scores, and probabilities — skills that are essential for success in the Applications and Interpretation as well as the Analysis and Approaches courses.
在IB数学中，练习18F通常会要求你运用正态分布的性质分析现实数据、通过标准化方法计算概率，并对总体做出推断。这一练习整合了你处理连续随机变量的能力，让你熟练使用反向正态函数，并理解正态模型适用的条件。掌握18F意味着你能在原始数据、z分数和概率之间自如转换——这些技能对于在《应用与解释》和《分析与方法》两门课程中取得成功都至关重要。
1. The Normal Distribution as a Continuous Model | 作为连续模型的正态分布
A continuous random variable X follows a normal distribution with mean μ and variance σ², denoted X ~ N(μ, σ²). The probability density function is symmetric about the mean, bell‑shaped, and the total area under the curve equals 1. Because the variable is continuous, the probability that X takes any single exact value is zero; we always compute probabilities over intervals using cumulative distribution functions.
如果一个连续随机变量X服从均值为μ、方差为σ²的正态分布，记作 X ~ N(μ, σ²)。概率密度函数关于均值对称，呈钟形，曲线下总面积为1。由于变量是连续的，X取任意单一精确值的概率为零；我们始终使用累积分布函数计算区间内的概率。
2. The Standard Normal Variable Z | 标准正态变量Z
To compare different normal distributions, we standardise any observation x by subtracting the mean and dividing by the standard deviation: z = (x − μ) ÷ σ. The resulting variable Z follows the standard normal distribution N(0, 1), which has mean 0 and standard deviation 1. Standardising allows us to use a single set of probability tables or calculator functions regardless of the original parameters.
为了比较不同的正态分布，我们通过减去均值并除以标准差将任意观测值x进行标准化：z = (x − μ) ÷ σ。由此得到的变量Z服从标准正态分布 N(0, 1)，其均值为0、标准差为1。标准化使我们能够只使用一套概率表或计算器函数，而不必考虑原始参数。
3. Using the Normal Cumulative Distribution Function | 使用正态累积分布函数
To find P(X < a) given X ~ N(μ, σ²), you can either standardise to Z and look up Φ(z) on a standard normal table, or use your GDC's normalcdf function. Normalcdf requires a lower bound, an upper bound, the mean, and the standard deviation. For an unbounded left tail, use a very small number like −10⁹⁹; for an unbounded right tail, use 10⁹⁹. Always sketch the curve and shade the region of interest to avoid direction errors.
已知X ~ N(μ, σ²)，要求出P(X < a)，你既可以将其标准化为Z然后查标准正态表Φ(z)，也可以使用图形计算器中的normalcdf函数。Normalcdf需要输入下限、上限、均值和标准差。对于无界的左尾，使用非常小的数值如 −10⁹⁹；对于无界的右尾，使用10⁹⁹。始终画出示意图并阴影标记目标区域，以避免方向错误。
4. Inverse Normal Calculations: From Probability to Value | 反向正态计算：从概率到取值
If a probability is given and you need the corresponding x‑value, use the inverse normal function invNorm (or InvNormal) on your GDC. This function asks for the area to the left of the unknown value, the mean, and the standard deviation. For symmetrical intervals around the mean, remember that the tail areas are (1 − confidence level) / 2. Understanding the direction of the area is critical — invNorm always returns the value that accumulates the specified left‑hand tail probability.
如果已知概率而需要求出对应的x值，请使用图形计算器上的invNorm（或InvNormal）函数。该函数需要输入未知值左侧的面积、均值和标准差。对于关于均值对称的区间，记住尾部面积为 (1 − 置信水平) / 2。理解面积的方向至关重要——invNorm始终返回累计左侧尾部概率达到指定值时对应的x值。
5. Non‑Standard Distributions and Calculator Tips | 非标准分布与计算器使用技巧
Modern IB exams expect you to use a GDC for most normal probability work. Familiarise yourself with the menus: on TI‑Nspire, use Menu > Statistics > Distributions > Normal Cdf or Inverse Normal; on Casio, navigate to STAT, then DIST, then NORM. Always double‑check that you have entered the standard deviation, not the variance, and that the bounds are in the correct order. For problems involving the sample mean, remember to adjust the standard deviation to σ ÷ √n.
现代IB考试期望你使用图形计算器完成大多数正态概率运算。请熟悉操作菜单：在TI‑Nspire上，依次进入Menu > Statistics > Distributions > Normal Cdf或Inverse Normal；在Casio上，则导航至STAT，再选择DIST，然后选择NORM。务必仔细检查你输入的是标准差而非方差，并且上下限顺序正确。对于涉及样本均值的问题，记得将标准差调整为 σ ÷ √n。
6. Real‑World Contexts in Exercise 18F | 练习18F中的现实背景
Exercise 18F problems are often built around scenarios such as manufacturing tolerances, human heights, examination scores, or packaging weights. You will be asked to find the proportion of items meeting a specification, determine a cut‑off point that excludes a given percentage, or compare two different normal distributions. Translating the verbal constraints into mathematical inequalities — e.g. 'between 45 kg and 55 kg' becomes P(45 < X < 55) — is a skill that needs deliberate practice.
练习18F的题目通常围绕制造公差、人体身高、考试成绩或包装重量等场景构建。你会被要求找出符合规格的产品比例、确定排除指定百分比的截断点，或者比较两个不同的正态分布。将文字约束转化为数学不等式——例如“介于45千克到55千克之间”转化为 P(45 < X < 55)——是一项需要刻意练习的技能。
7. The 68–95–99.7 Rule and Quick Estimates | 68–95–99.7规则与快速估计
The empirical rule states that about 68% of data lie within 1 standard deviation of the mean, 95% within 2 standard deviations, and 99.7% within 3 standard deviations. While you will use precise calculator values in exams, this rule helps you sanity‑check your answers rapidly. If a question asks for the proportion of outcomes more than 2σ below the mean, you expect roughly 2.5% — if your GDC returns 0.023, you know it is plausible.
经验法则指出，大约68%的数据落在均值左右1个标准差之内，95%落在2个标准差之内，99.7%落在3个标准差之内。虽然在考试中你会使用精确的计算器数值，但这一法则可以帮助你快速检验答案的合理性。如果题目问及低于均值超过2σ的结果占比，你期望大概为2.5%——如果你的图形计算器返回0.023，你就知道这个结果是合理的。
8. Conditions for Using a Normal Model | 使用正态模型的条件
Not every data set can be modelled by a normal distribution. The underlying variable should be continuous, unimodal, roughly symmetric, and free from extreme outliers. In IB problems, you may be given a histogram or a box‑plot and asked whether a normal model is appropriate. Look for skewness, gaps, or heavily truncated data. A normal probability plot (Q‑Q plot) can also be used: if the points roughly follow a straight line, normality is a reasonable assumption.
并非所有数据集都能用正态分布进行建模。底层变量应当是连续的、单峰的、近似对称的，并且没有极端异常值。在IB问题中，你可能会拿到直方图或箱线图，并需要判断正态模型是否适用。请留意偏斜、缺口或严重截断的数据。正态概率图（Q‑Q图）也可用于判断：如果各点大致沿一条直线排列，则正态性是一个合理的假设。
9. The Distribution of the Sample Mean | 样本均值的分布
When we take repeated samples of size n from a normal population, the sample mean X̄ also follows a normal distribution with mean μ and standard deviation σ ÷ √n. Even if the population is not normal, the Central Limit Theorem assures that X̄ his fact is central to hypothesis testing and confidence intervals, which often follow directly from Exercise 18F concepts.
当我们从正态总体中反复抽取容量为n的样本时，样本均值 X̄ 也服从正态分布，其均值为 μ，标准差为 σ ÷ √n。即使总体不服从正态分布，中心极限定理也能保证当n足够大（通常n ≥ 30）时，X̄ 近似服从正态分布。这一事实正是假设检验和置信区间的核心，而这些内容往往直接承接练习18F的概念。
10. Normal Approximation to the Binomial Distribution | 二项分布的正态近似
ity correction: for P(X ≤ k) use P(X < k + 0.5) in the normal model. This improves accuracy for discrete‑to‑continuous transitions.
如果一个二项随机变量的参数 n 和 p 满足 np ≥ 5 且 n(1 − p) ≥ 5，那么该分布可用均值为 np、标准差为 √(np(1 − p)) 的正态分布来近似。在练习18F中，你可能会被要求使用这一近似来更快地计算概率，但务必记得应用连续性校正：对于 P(X ≤ k)，在正态模型中应使用 P(X < k + 0.5)。这能提高离散到连续过渡时的准确性。
11. Step‑by‑Step Solution to a Classic 18F Problem | 典型18F题目的分步解析
Let us solve a modelled question: ' eviation 0.08 kg. Find the probability that a randomly chosen bag weighs more than 1.2 kg.' 304. Alternatively, calculate directly: normalcdf(1.2, 10⁹⁹, 1.05, 0.08). Thus, about 3.04% of bags weigh more than 1.2 kg. Always interpret the result in context.
我们来解决一个模拟题目：“一袋袋苹果的质量服从正态分布，均值为1.05 kg，标准差为0.08 kg。求随机选取一袋苹果其质量超过1.2 kg的概率。”首先确定 μ = 1.05，σ = 0.08。所求概率为 P(X > 1.2)。标准化：z = (1.2 − 1.05) ÷ 0.08 = 1.875。使用图形计算器：normalcdf(1.875, 10⁹⁹, 0, 1) ≈ 0.0304。或者直接计算：normalcdf(1.2, 10⁹⁹, 1.05, 0.08)。因此，大约3.04%的袋子重量超过1.2 kg。务必结合背景解释这一结果。
12. Common Pitfalls and How to Avoid Them | 常见陷阱与规避方法
The most frequent error is swapping the mean and standard deviation, or entering the variance instead of the standard deviation into the calculator. Another trap is misreading 'at least' as 'greater than' without accounting for equality in a discrete approximation, or forgetting to halve the significance level in a two‑tailed inverse normal problem. Finally, always differentiate between a question about a single observation and one about a sample mean — the latter uses the standard error σ ÷ √n, which dramatically changes the result.
最常见的错误是混淆均值和标准差，或将方差而非标准差输入计算器。另一个陷阱是在离散近似问题中，将“至少”错误理解为“大于”而忽略了等号的处理，或者在双尾反向正态问题中忘记将显著性水平减半。最后，务必区分针对单个观测值的问题和针对样本均值的问题——后者要使用标准误差 σ ÷ √n，这会显著改变结果。
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