The Signal and the Noise
Also known as The Signal and the Noise: Why So Many Predictions Fail – but Some Don't, The Signal and the Noise: The Art and Science of Prediction
Nate Silver's 2012 book on using probability and statistics to make better predictions.
At a glance
Key ideas
Calibration, Say the odds, Bayesian updating …
See all 4Criticism
Michael E. Mann
Connections
Also on this page: Key facts · Links · Sources · Related pages
- Silver describes the practical art of building models with probability and statistics.
- Case studies cover baseball, elections, climate change, the 2008 financial crisis, poker and weather forecasting.
- He asks for uncertainty to be stated as ranges and probabilities, not single point estimates.
- He argues for a Bayesian approach of updating probabilities as new data arrives, over the frequentist methods of Ronald Fisher.
Why it matters
- It explains how forecasting works without the mathematics, and its sales jumped after the 2012 US election that Silver's model called.
Key facts
- Author
- Nate Silver1
- Published
- 27 Sep 2012, United States1
- UK edition
- April 2013, as The Signal and the Noise: The Art and Science of Prediction (Allen Lane; Penguin paperback)1
- US paperback
- 3 Feb 2015, with a new introduction1
- Translations
- Including Portuguese, German, Italian, Finnish, Chinese, Japanese, Spanish, Romanian, Czech and Korean1
- Award
- Phi Beta Kappa Society book award in science, 20131
Key ideas
-
Calibration1
Weather forecasting shows how well stated probabilities match what actually happens.
-
Say the odds1
Replace "a toss-up state" with probabilities, such as an 83% chance of winning.
-
Bayesian updating1
Use long-run data and move probabilities up or down as new evidence comes in.
-
Know the domain first1
A statistician first needs to understand how the process works and which measures are reliable.
Criticism
-
Michael E. Mann1
Criticised it for treating physical climate trends like the social phenomena of voter preferences.
Connections
Written by
Links
Sources
Last checked against its sources today
Is something on this page wrong or out of date? Tell us
The line under a date fills as time passes: empty when new, then one equal share each for a day, a week, a month and a year. For episodes and other items it is full at 2 years; for "last checked", at a year. A striped line means the date is known only to the month or year.