How (not) to forecast an election: Analysis via a hierarchical Bayesian model

Re:How (not) to forecast an election: Analysis via a hierarc... by whybroke on Sunday November 27, 2016 @04:53PM (#13051775)

Actually the discrepancy stems from underestimating voter turnout for Trump and over estimating it for Clinton and is not caused by the mythical timid Trump voter.

[ Parent | on HN ]

Re:How (not) to forecast an election: Analysis via a hierarc... by imh on Monday November 28, 2016 @11:47AM (#13056591)

Doesn't matter. All that matters is that the errors are correlated. If voter turnout is biased in one state, then it likely is in another. The reluctance was just an example.

[ Parent | on HN ]