Re:Nate Silver correctly called 50 out of 50 states
by 3JPLW on Wednesday November 07, 2012 @12:53AM
(#4752359)
His models have a memory (increase sample size by using a handful of day or week old data) and aggregate as many as 5 to 8 independent polls per state plus as many as a dozen national polls that factor into his simulations. That's easily four to sixteen times the sample size.
Re:Nate Silver correctly called 50 out of 50 states
by niggler on Wednesday November 07, 2012 @12:55AM
(#4752367)
You can improve effective sample size by eliminating extraneous variables. One thing that Nate does is take advantage of implicit biases (for example, PPP polls lean slightly left)
Re:Nate Silver correctly called 50 out of 50 states
by waterlesscloud on Wednesday November 07, 2012 @05:34AM
(#4752947)
PPP seems to have predicted 332 EV and correctly picked the winner in every state they polled. They didn't poll some of the obvious states...
How does this work? I was under the impression that to shrink the confidence interval by half you need 4x as many sample points, but there must be more sophisticated statistical methods at work here.