|
Activity Number:
|
312
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Social Statistics Section
|
| Abstract - #301839 |
|
Title:
|
Predicting Electoral Vote Totals for the Presidential Election
|
|
Author(s):
|
Christopher J. Rigdon*+ and Steven Rigdon and Edward Sewell and Sheldon Jacobsen
|
|
Companies:
|
Arizona State University and Southern Illinois University Edwardsville and Southern Illinois University Edwardsville and University of Illinois at Urbana-Champaign
|
|
Address:
|
944 S. Terrace Rd. #204, Tempe, AZ, 85281,
|
|
Keywords:
|
presidential election ; polling data ; Bayesian statistics
|
|
Abstract:
|
During the U.S. Presidential Race of 2008, there will be hundreds of polls conducted in each state to determine which candidate is most likely to win the election at any given time. The objective of this research is to use polling data to derive an empirical posterior distribution for the number of electoral votes each candidate will receive. By expanding on the work of Kaplan and Barnett (2003), a model is derived which considers undecided voters which can have a significant effect on election day. The model performed quite well in the 2004 election, missing the correct number of electoral votes for each candidate by only two.
|
