Jump to content

File:Statistical bias and statistical noise illustration.png

Page contents not supported in other languages.
This is a file from the Wikimedia Commons
From Wikipedia, the free encyclopedia

Original file (748 × 830 pixels, file size: 118 KB, MIME type: image/png)

Summary

Description
English: The targets and shots in this figure illustrate the difference between statistical bias and statistical noise and how these concepts relate to statistical error. The shots on target b) are biased, because the shots are systematically off in one direction (down to the right from the bullseye). On average, there is an error. Target c) on the other hand does not have a mean error, because the imprecision in each shot in relation to the bullseye cancel each other out. So there is no bias – but there is more noise, because the shots there differ much from each other. The shots on target a) are precise; there is no bias and very little noise. Target d) has it worst, since it suffers from both bias and noise. A final observation here is that if there were no targets, it would be impossible to say that the a) or b) shots are biased, since there would be no true value (the bullseye) to compare them to, but it would still be obvious that the c) and d) shots are noisy. (Adapted from Kahneman, D., Sibony, O. & Sunstein, C. (2021). Noise: A flaw in human judgment. New York: Little, Brown Spark. ISBN 978-0-00-830899-5. OCLC 1242782025.)
Svenska: Måltavlorna och skotten i denna figur illustrerar skillnaden mellan statistisk bias och statistiskt brus och hur dessa koncept förhåller sig till statistiskt fel. Skotten på måltavla b) har bias, för skotten har en systematisk förskjutning i en riktning (ner till höger från bullseye). I genomsnitt finns det ett statistiskt fel. Måltavla c) däremot har inget statistiskt fel i medelvärde, eftersom varje skotts förskjutning i förhållande till bullseye tar ut varandra. Så tavlan har ingen bias – men den har mer brus, för skotten där skiljer sig mycket åt från varandra. Skotten på måltavla a) är träffsäkra; där finns ingen bias och väldigt lite brus. Måltavla d) har det värst, eftersom den lider av både bias och brus. En sista observation här är att om det inte fanns några måltavlor så skulle det vara omöjligt att säga att a)- och b)-skotten har bias eller inte, eftersom det inte skulle finnas något sant värde (bullseye) att att jämföra dem mot, men det skulle fortfarande vara uppenbart att c)- och d)- skotten är brusiga. (Anpassad från Kahneman, D., Sibony, O. & Sunstein, C. (2021). Brus: Det osynliga felet som stör våra bedömningar – och vad du kan göra åt det. Stockholm: Volante. ISBN 978-91-7965-105-3)
Date
Source Own work
Author Tim Isaksson

Licensing

I, the copyright holder of this work, hereby publish it under the following license:
Creative Commons CC-Zero This file is made available under the Creative Commons CC0 1.0 Universal Public Domain Dedication.
The person who associated a work with this deed has dedicated the work to the public domain by waiving all of their rights to the work worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law. You can copy, modify, distribute and perform the work, even for commercial purposes, all without asking permission.

Captions

A target-analogy illustration of statistical bias (b), statistical noise (c), both (d) and none (accuracy; a). Adapted from Kahneman, Sibony & Sunstein (2021).

Items portrayed in this file

depicts

30 June 2021

File history

Click on a date/time to view the file as it appeared at that time.

Date/TimeThumbnailDimensionsUserComment
current13:52, 30 June 2021Thumbnail for version as of 13:52, 30 June 2021748 × 830 (118 KB)Tim IsakssonUploaded own work with UploadWizard

The following 2 pages use this file:

Global file usage

The following other wikis use this file:

Metadata