计算正态样本的期望极差
本文深入探讨了如何计算来自标准正态分布 N(0, 1) 的 n 个样本的期望极差(最大值与最小值之差)。结果以 σ 为单位给出,若 σ 不为 1,则需乘以相应 σ 值。文章基于此前关于12人陪审团中智商期望极差的讨论,提供了更通用的计算方法。
本文深入探讨了如何计算来自标准正态分布 N(0, 1) 的 n 个样本的期望极差(最大值与最小值之差)。结果以 σ 为单位给出,若 σ 不为 1,则需乘以相应 σ 值。文章基于此前关于12人陪审团中智商期望极差的讨论,提供了更通用的计算方法。
In a randomly selected jury of 12 people, the expected highest IQ is about 117 and the lowest about 83, giving a spread of roughly 34 points, based on a normal distribution with mean 100 and standard deviation 15.
The article discusses how to calculate the expected range for samples from a normal distribution, explaining that the expected range depends on sample size and is expressed as a constant multiple of the standard deviation. It presents tables and approximations for these constants, known as d₂ values, which are used in quality control and statistical process control.