The Normal Distribution | 正态分布

📚 The Normal Distribution | 正态分布

The normal distribution is one of the most important continuous probability distributions in A-Level Mathematics. It models variables such as heights, weights, measurement errors and exam marks, where most observations cluster around a central mean and extreme values become less likely. Understanding its shape, parameters and standardisation process is essential for Edexcel Statistics questions.

正态分布是 A-Level 数学中最重要的连续概率分布之一。它用于描述身高、体重、测量误差和考试成绩等变量,这些数据大多集中在均值附近,极端值出现的概率较低。理解正态曲线的形状、参数以及标准化过程是解答 Edexcel 统计题的关键。


1. Definition and Continuous Nature | 定义与连续性

A continuous random variable X follows a normal distribution with mean μ and variance σ² if its probability density function is symmetric, bell-shaped and defined for all real x. We write X ~ N(μ, σ²). Note that the second parameter is variance, not standard deviation.

如果连续随机变量 X 服从均值为 μ、方差为 σ² 的正态分布,则其概率密度函数呈对称钟形,且对所有实数 x 有定义。我们记作 X ~ N(μ, σ²)。请注意,第二个参数是方差,而不是标准差。

f(x) = 1/(σ√(2π)) × e−½((x−μ)/σ)²

The total area under the probability density curve is exactly 1, because the normal distribution is a continuous probability distribution. This means the curve never gives a probability at a single point; instead, probabilities are found as areas under the curve over an interval.

概率密度曲线下的总面积恰好为 1,因为正态分布是一种连续概率分布。这意味着曲线在单点处并不给出概率;相反,概率表示为一个区间内曲线下的面积。


2. Shape and Symmetry | 形状与对称性

The normal curve is perfectly symmetrical about the vertical line x = μ. This implies P(X < μ) = P(X > μ) = 0.5. The curve has a single peak at the mean, and its tails extend indefinitely in both directions but never touch the horizontal axis.

正态曲线关于直线 x = μ 完全对称。因此 P(X < μ) = P(X > μ) = 0.5。曲线在均值处只有一个峰值,两端尾部无限延伸但永远不会接触横轴。

  • Mean = median = mode for a normal distribution. | 正态分布的均值、中位数和众数相等。
  • The total area under the curve equals 1. | 曲线下总面积为 1。
  • The x-axis is a horizontal asymptote. | x 轴是水平渐近线。

Because of symmetry, probabilities on either side of the mean mirror each other. For example, if P(X < a) = p, then P(X > 2μ − a) = p as well. This property often allows exam questions to be simplified.

由于对称性,均值两侧的概率互为镜像。例如,如果 P(X < a) = p,那么 P(X > 2μ − a) 也等于 p。这一性质常常可以简化考题。


3. Parameters μ and σ | 参数 μ 与 σ

The mean μ determines the centre of the distribution. Changing μ shifts the entire curve left or right without changing its shape. The standard deviation σ controls the spread; a larger σ gives a flatter and wider curve, while a smaller σ gives a taller and narrower curve.

均值 μ 决定了分布的中心位置。改变 μ 只会使整条曲线左右平移,而不会改变形状。标准差 σ 控制离散程度;σ 越大,曲线越平坦宽阔;σ 越小,曲线越高窄。

In Edexcel notation, X ~ N(μ, σ²) always shows the variance as the second parameter. If a question says X ~ N(100, 225), then μ = 100 and σ² = 225, so σ = 15. You must identify this correctly before using a calculator.

在 Edexcel 记号中,X ~ N(μ, σ²) 始终用方差作为第二个参数。如果题目给出 X ~ N(100, 225),则 μ = 100,σ² = 225,因此 σ = 15。在使用计算器之前必须正确识别这一点。


4. The 68-95-99.7 Rule | 68-95-99.7 规则

For any normal distribution, approximately 68% of observations lie within one standard deviation of the mean, 95% lie within two standard deviations, and 99.7% lie within three standard deviations. These percentages are useful for quick estimates but exact probabilities should be found using a calculator or tables.

对于任何正态分布,大约 68% 的观测值落在均值的一个标准差范围内,95% 落在两个标准差范围内,99.7% 落在三个标准差范围内。这些百分比可用于快速估计,但精确概率应使用计算器或表格求出。

Interval Approximate probability
μ ± 1σ 0.68
μ ± 2σ 0.95
μ ± 3σ 0.997

This rule is often tested in multiple-choice or data-interpretation questions where a quick judgement about outliers or usual values is needed.

该规则常用于选择题或数据解释题,需要快速判断异常值或正常取值范围。


5. Standardising to Z | 标准化为 Z

To find probabilities for any normal variable X ~ N(μ, σ²), we convert X to the standard normal variable Z using Z = (X − μ)/σ. The standard normal distribution has mean 0 and variance 1, written Z ~ N(0, 1). Standardisation allows all normal probabilities to be compared on the same scale.

为了求任意正态变量 X ~ N(μ, σ²) 的概率,我们使用 Z = (X − μ)/σ 将 X 转换为标准正态变量 Z。标准正态分布的均值为 0,方差为 1,记作 Z ~ N(0, 1)。标准化使得所有正态概率可以在同一尺度上比较。

Z = (X − μ) / σ

For example, if X ~ N(70, 25) and you want P(X < 74), first find σ = 5. Then Z = (74 − 70)/5 = 0.8, so P(X < 74) = P(Z < 0.8). This is the value you look up or obtain from the calculator.

例如,若 X ~ N(70, 25),要求 P(X < 74),首先得到 σ = 5。然后 Z = (74 − 70)/5 = 0.8,因此 P(X < 74) = P(Z < 0.8)。这就是你要查表或从计算器获得的值。


6. Calculating Probabilities | 计算概率

Edexcel questions often ask for P(X < a), P(X > a) or P(a < X < b). Because the normal distribution is continuous, P(X < a) = P(X ≤ a) and P(X > a) = P(X ≥ a). Use the symmetry property and the complementary rule P(Z > z) = 1 − P(Z < z) when necessary.

Edexcel 题目常要求计算 P(X < a)、P(X > a) 或 P(a < X < b)。由于正态分布是连续分布,P(X < a) = P(X ≤ a),并且 P(X > a) = P(X ≥ a)。必要时使用对称性和互补规则 P(Z > z) = 1 − P(Z < z)。

On most approved calculators, you can enter the lower bound, upper bound, μ and σ directly to find P(a < X < b). For one-tailed probabilities, use a very small lower bound such as −10⁹ or a very large upper bound such as 10⁹ to approximate the infinite tail.

在大多数考试允许的计算器上,你可以直接输入下界、上界、μ 和 σ 来求 P(a < X < b)。对于单尾概率,可使用一个很小的下界(如 −10⁹)或很大的上界(如 10⁹)来近似无限尾部。

  • P(X < a): lower = −10⁹, upper = a. | P(X < a):下界 −10⁹,上界 a。
  • P(X > a): lower = a, upper = 10⁹. | P(X > a):下界 a,上界 10⁹。
  • P(a < X < b): lower = a, upper = b. | P(a < X < b):下界 a,上界 b。

7. Inverse Normal Calculations | 正态反查

If a probability is given and you need to find the corresponding boundary value, use the inverse normal function on your calculator. For example, to find k such that P(X < k) = 0.95, enter area 0.95, mean μ and standard deviation σ into the inverse normal function.

如果已知概率,需要求对应的临界值,可使用计算器上的正态反查功能。例如,要求满足 P(X < k) = 0.95 的 k,在反查功能中输入面积 0.95、均值 μ 和标准差 σ 即可。

For right-tail problems, convert the condition first. If P(X > k) = 0.05, then P(X < k) = 0.95. If X ~ N(100, 15²), then k ≈ 124.7. This means only the top 5% of values are above 124.7.

对于右尾问题,要先转化条件。如果 P(X > k) = 0.05,则 P(X < k) = 0.95。若 X ~ N(100, 15²),则 k ≈ 124.7。这意味着只有最高的 5% 数值大于 124.7。

The inverse normal function always works with left-tail cumulative probabilities. Writing the correct left-tail probability is the most common skill required in exam questions.

正态反查功能始终使用左尾累积概率。写出正确的左尾概率是考试题中最常考查的技能。


8. Finding Unknown μ and σ | 求解未知 μ 和 σ

Some exam questions provide two probability statements and ask you to find unknown μ and σ. Write each statement in standardised form, for example P(X < a) = p becomes P(Z < (a−μ)/σ) = p. Then use the inverse normal Z-value and solve the simultaneous equations.

有些考题会给出两个概率条件,要求求解未知的 μ 和 σ。将每个条件写成标准化形式,例如 P(X < a) = p 变为 P(Z < (a−μ)/σ) = p。然后使用反查得到的 Z 值,建立并解联立方程。

Suppose X ~ N(μ, σ²), P(X < 50) = 0.2 and P(X < 60) = 0.7. The corresponding Z-values are approximately −0.8416 and 0.5244. This gives the simultaneous equations 50 = μ − 0.8416σ and 60 = μ + 0.5244σ. Solving gives σ ≈ 7.32 and μ ≈ 56.16.

假设 X ~ N(μ, σ²),P(X < 50) = 0.2 且 P(X < 60) = 0.7。对应的 Z 值约为 −0.8416 和 0.5244。由此得到联立方程 50 = μ − 0.8416σ 和 60 = μ + 0.5244σ。解得 σ ≈ 7.32,μ ≈ 56.16。

Always keep the Z-values to at least four significant figures during calculation. Rounding too early can produce inaccurate final answers, especially when the equations are close together.

计算过程中 Z 值至少保留四位有效数字。过早舍入可能导致最终答案不准确,尤其是当两个方程比较接近时。


9. Normal Approximation to Binomial | 二项分布的正态近似

When n is large and p is not too close to 0 or 1, a binomial distribution B(n, p) can be approximated by a normal distribution with μ = np and σ² = np(1−p). The usual Edexcel condition is that both np and nq are greater than 5 (some texts use 10).

当 n 较大且 p 不太接近 0 或 1 时,二项分布 B(n, p) 可用均值为 np、方差为 np(1−p) 的正态分布近似。Edexcel 通常要求 np 和 nq 都大于 5(有些教材使用 10)。

If X ~ B(n, p) and np > 5, nq > 5, then X ~ N(np, npq) approximately.

Here q = 1 − p. This approximation is useful when n is too large for exact binomial calculations. In exam questions you should state the approximation, justify the conditions, and then apply a continuity correction.

这里 q = 1 − p。当 n 太大而难以进行精确的二项计算时,这种近似非常有用。在考试题中,你应写出近似、验证条件,然后进行连续性校正。


10. Continuity Correction | 连续性校正

Since binomial is discrete and normal is continuous, we apply a continuity correction. Replace P(X = a) with P(a−0.5 < X < a+0.5), P(X ≤ a) with P(X < a+0.5), and P(X ≥ a) with P(X > a−0.5) in the normal approximation.

由于二项分布是离散的,而正态分布是连续的,我们需要进行连续性校正。在正态近似中,用 P(a−0.5 < X < a+0.5) 代替 P(X = a),用 P(X < a+0.5) 代替 P(X ≤ a),用 P(X > a−0.5) 代替 P(X ≥ a)。

Discrete event Normal approximation
P(X = a) P(a−0.5 < X < a+0.5)
P(X ≤ a) P(X < a+0.5)
P(X ≥ a) P(X > a−0.5)
P(X < a) P(X < a−0.5)
P(X > a) P(X > a+0.5)

Forgetting the continuity correction is a common error. Even if the question does not explicitly say ‘use a normal approximation’, you must include the correction whenever a discrete binomial is approximated by a continuous normal distribution.

忘记连续性校正是常见错误。即使题目没有明确说“使用正态近似”,只要用连续的正态分布近似离散的二项分布,就必须进行校正。


11. Exam Tips and Common Errors | 考试技巧与常见错误

Always check whether the given parameter is variance or standard deviation. If X ~ N(100, 225), then σ² = 225 and σ = 15. Many students incorrectly use 225 as σ. Also remember that the calculator needs σ, not σ², for normal and inverse normal functions.

务必检查题目给出的参数是方差还是标准差。如果 X ~ N(100, 225),则 σ² = 225,σ = 15。许多学生错误地将 225 当作 σ。此外,计算器中的正态分布和反查功能需要输入 σ,而不是 σ²。

Comments

屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导,国外大学本科硕士研究生博士课程论文辅导

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Discover more from aleveler.com

Subscribe now to keep reading and get access to the full archive.

Continue reading