[position in research by date]
[position in research by subject]
"Interpreting the Predictive Uncertainty of
Elections,"
Journal of Politics, April 2009, 612-626.
pdf file (1,448 KB).
Abstract
This paper provides an interpretation of the uncertainty that exists
on election morning as to who will win.
It is based on the theory that there are a number of possible conditions
of nature that can exist on election day, of which one is drawn.
Political betting markets
provide a way of trying to estimate this uncertainty.
This type of uncertainty is different from polling standard errors, which
estimate sample-size uncertainty.
This paper also introduces a ranking assumption concerning dependencies
across U.S. states. The assumption
does well when tested using data for the
2004 presidential election and the 2006 Senate election.
It is shown that if the ranking assumption is correct, the two political
parties should spend all their money on a few states, which seems
consistent with their actual behavior in 2004.
Comments
This paper is a substantially revised version of an earlier paper
entitled "Predicting Electoral College Victory Probabilities from
State Probability Data," November 2004.
The ranking assumption, while simple, looks really good!