Lesson 6 of 81908–19377 min
Confidence intervals
William Sealy Gosset worked as a chemist at the Guinness brewery in Dublin, where experiments on barley and yeast came in small batches. The normal-based methods of the day assumed large samples and broke down with five or ten observations. Guinness let him publish on condition that he used a pen name, so his 1908 paper in Biometrika is signed “Student”, and the distribution he found is still Student's t.
In 1937 Jerzy Neyman gave interval estimation its modern form, the confidence interval, together with the precise, slightly awkward meaning of the word confidence that students still have to learn.
The rest of this lesson is part of Pro
Create a free account to read the first lesson of every path, and go Pro for the rest.
Sources
- Outline of a Theory of Statistical Estimation Based on the Classical Theory of Probability (J. Neyman, Phil. Trans. R. Soc. A, 1937) — The Royal Society
- The Probable Error of a Mean (“Student”, Biometrika 6(1), 1908) — Biometrika (Oxford University Press)
- NIST/SEMATECH e-Handbook of Statistical Methods: Confidence Limits for the Mean — NIST