Difference between revisions of "Talk:2136: Election Commentary"
(Created by dgbrtBOT) |
(Bayes?) |
||
| Line 1: | Line 1: | ||
<!--Please sign your posts with ~~~~ and don't delete this text. New comments should be added at the bottom.--> | <!--Please sign your posts with ~~~~ and don't delete this text. New comments should be added at the bottom.--> | ||
| + | A lot of election commentary falls under the category of Bayesian inference. The various news agencies have prior distributions for the votes from all of the precincts, and update their estimates of the final total votes based on the precincts that have reported so far. Thus, "Candidate X is leading Candidate Y in votes received, but since only the precincts that previously have favoured X-like candidates have reported in so far, and they aren't giving X as much a lead as expected, we think this means Y will win overall." [[Special:Contributions/162.158.63.208|162.158.63.208]] 18:11, 12 April 2019 (UTC) | ||
Revision as of 18:11, 12 April 2019
A lot of election commentary falls under the category of Bayesian inference. The various news agencies have prior distributions for the votes from all of the precincts, and update their estimates of the final total votes based on the precincts that have reported so far. Thus, "Candidate X is leading Candidate Y in votes received, but since only the precincts that previously have favoured X-like candidates have reported in so far, and they aren't giving X as much a lead as expected, we think this means Y will win overall." 162.158.63.208 18:11, 12 April 2019 (UTC)
