27 February 2020, 10:36:11UTC RSS | About | Contact | Site Map
Home » News & Blogs » So what’s all the fuss about Bayesian statistics?
Galileo's Pendulum

## So what’s all the fuss about Bayesian statistics?

7 Jun 2013, 17:39 UTC
(200 words excerpt, click title or image to see full post)

When someone thinks sexy science, a statistical method doesn’t usually come to mind. However, Nate Silver — a baseball statistician who made a name for himself by analyzing political polls for more than four years — made people who usually shun such topics aware of a technique known as Bayesian inference. Silver used Bayesian methods […]

This is probably not Thomas Bayes, but it’s the picture we all use for him anyway. [Source: Wikipedia]When someone thinks sexy science, a statistical method doesn’t usually come to mind. However, Nate Silver — a baseball statistician who made a name for himself by analyzing political polls for more than four years — made people who usually shun such topics aware of a technique known as Bayesian inference. Silver used Bayesian methods to combine the results of various polls, analyze them, and predict the probability of a given candidate winning the majority of votes in a district. In each race (presidential and senatorial) in the 2012 election in the United States, the candidate he listed as most likely to win did.
The key to Silver’s success was twofold. He assumed (shockingly enough) that political polls in general are largely accurate, based on the outcomes ...