题目
Problem
Components are manufactured such that their length in mm is normally distributed with mean μ and variance σ2. Below is a 95% confidence interval for μ calculated from a random sample of components.
(11.52, 13.75)
Using the same random sample,
(a) find a 90% confidence interval for μ.
(6)
Four 90% confidence intervals are found from independent random samples.
(b) Calculate the probability that only 3 of these 4 intervals will contain μ.
(2)
(Total for Question 3 is 8 marks)
题目中文翻译
元件的长度(单位 mm)服从均值为 μ、方差为 σ2 的正态分布。下面是根据一个元件样本算出的 μ 的 95% 置信区间。
(11.52, 13.75)
使用同一个样本,
(a) 求 μ 的 90% 置信区间。
(b) 从相互独立的 4 个随机样本中得到 4 个 90% 置信区间。求其中恰好只有 3 个区间包含 μ 的概率。
解答
(a)
解法一
思路
展开
同一个样本给出的不同置信水平区间具有相同的样本均值与标准误。先由原 95% 区间的中点求 xˉ,再用半宽除以 1.96 求标准误;最后改用 90% 区间的临界值 1.6449。
答题过程
展开
The sample mean is the midpoint of the 95% confidence interval:
xˉ==211.52+13.7512.635.
The half-width of the 95% confidence interval is
13.75−12.635=1.115.
Since the 95% critical value is 1.96, the standard error is
SE==1.961.1150.568877…
For a 90% confidence interval, the critical value is 1.6449. Hence the interval is
===xˉ±1.6449SE12.635±1.6449(0.568877…)12.635±0.9357…(11.699…, 13.570…).
Therefore, to 3 significant figures, the 90% confidence interval is
(11.7, 13.6) mm.
(b)
解法一
思路
展开
每个 90% 置信区间包含 μ 的概率为 0.9,不包含的概率为 0.1。四个区间来自独立样本,因此包含 μ 的区间数服从 Bin(4,0.9),计算恰好成功三次的概率。
答题过程
展开
Let X be the number of the four confidence intervals that contain μ. Then
X∼Bin(4,0.9).
Therefore,
P(X=3)==≈(34)(0.9)3(0.1)0.29160.292.