×

zbMATH — the first resource for mathematics

Dynamic Bayesian forecasting of presidential elections in the States. (English) Zbl 06158330
Summary: I present a dynamic Bayesian forecasting model that enables early and accurate prediction of U.S. presidential election outcomes at the state level. The method systematically combines information from historical forecasting models in real time with results from the large number of state-level opinion surveys that are released publicly during the campaign. The result is a set of forecasts that are initially as good as the historical model, and then gradually increase in accuracy as Election Day nears. I employ a hierarchical specification to overcome the limitation that not every state is polled on every day, allowing the model to borrow strength both across states and, through the use of random-walk priors, across time. The model also filters away day-to-day variation in the polls due to sampling error and national campaign effects, which enables daily tracking of voter preferences toward the presidential candidates at the state and national levels. Simulation techniques are used to estimate the candidates’ probability of winning each state and, consequently, a majority of votes in the Electoral College. I apply the model to preelection polls from the 2008 presidential campaign and demonstrate that the victory of Barack Obama was never realistically in doubt.

MSC:
62 Statistics
Software:
R
PDF BibTeX XML Cite
Full Text: DOI
References:
[1] Abramowitz A. I., PS: Political Science & Politics 41 pp 691– (2008) · doi:10.1017/S1049096508081249
[2] Arceneaux K., British Journal of Political Science 36 pp 159– (2006) · doi:10.1017/S0007123406000081
[3] Bartels L. M., Political Science & Politics 34 pp 9– (2001) · doi:10.1017/S1049096501000026
[4] Becker B., The New York Times (2008)
[5] Biemer P. P., Handbook of Survey Research pp 27–, 2. ed. (2010)
[6] Broh C. A., Public Opinion Quarterly 44 pp 514– (1980) · doi:10.1086/268620
[7] DOI: 10.1080/10618600.1998.10474787 · doi:10.1080/10618600.1998.10474787
[8] Brown L. B., International Journal of Forecasting 15 pp 127– (1999) · doi:10.1016/S0169-2070(98)00059-4
[9] Campbell J. E., American Journal of Political Science 36 pp 386– (1992) · doi:10.2307/2111483
[10] Campbell J. E., American Politics Research 24 pp 408– (1996) · doi:10.1177/1532673X9602400402
[11] Campbell J. E., Political Science & Politics 34 pp 33– (2001) · doi:10.1017/S1049096501000051
[12] Campbell J. E., PS: Political Science & Politics 41 pp 679– (2008) · doi:10.1017/S1049096508081006
[13] Campbell J. E., PS: Political Science & Politics 41 pp 697– (2008) · doi:10.1017/S1049096508081158
[14] Campbell J. E., American Politics Research 20 pp 287– (1992) · doi:10.1177/1532673X9202000302
[15] Campbell J. E., Before the Vote: Forecasting American National Elections (2000)
[16] Campbell J. E., American Politics Research 18 pp 251– (1990) · doi:10.1177/1532673X9001800301
[17] Center for Responsive Politics. 2008. ”U.S. Election Will Cost \(5.3 Billion, Center for Responsive Politics Predicts". Available athttp://www.opensecrets.org/news/2008/10/us-election-will-cost-53-billi.html\)
[18] DOI: 10.1198/000313008X267820 · Zbl 05680762 · doi:10.1198/000313008X267820
[19] Dickerson J., Slate (2008)
[20] Duch R. M., The Economic Vote: How Political and Economic Institutions Condition Election Results (2008) · doi:10.1017/CBO9780511755934
[21] Erikson R. S., PS: Political Science and Politics 29 pp 37– (1996)
[22] Erikson R. S., Public Opinion Quarterly 63 pp 163– (1999) · doi:10.1086/297709
[23] Erikson R. S., PS: Political Science & Politics 41 pp 703– (2008) · doi:10.1017/S1049096508081237
[24] Fair R. C., American Journal of Political Science 53 pp 55– (2009) · doi:10.1111/j.1540-5907.2008.00357.x
[25] Farhi P., The Washington Post (2004)
[26] Finkel S. E., The Journal of Politics 55 pp 1– (1993) · doi:10.2307/2132225
[27] Gelman A., British Journal of Political Science 23 pp 409– (1993) · doi:10.1017/S0007123400006682
[28] Gelman A., Statistical Science 7 pp 457– (1992) · Zbl 1386.65060 · doi:10.1214/ss/1177011136
[29] Hillygus D. S., American Journal of Political Science 47 pp 583– (2003) · doi:10.1111/1540-5907.00041
[30] Holbrook T. M., American Journal of Political Science 35 pp 91– (1991) · doi:10.2307/2111439
[31] Holbrook T. M., American Journal of Political Science 38 pp 973– (1994) · doi:10.2307/2111729
[32] Holbrook T. M., International Journal of Forecasting 15 pp 137– (1999) · doi:10.1016/S0169-2070(98)00060-0
[33] DOI: 10.1080/10361140500302472 · doi:10.1080/10361140500302472
[34] Jamieson K. H. (ed.), Electing the President, 2008: The Insiders’ View (2009)
[35] Kaplan E. H., Operations Research 51 pp 32– (2003) · Zbl 1163.91542 · doi:10.1287/opre.51.1.32.12794
[36] Kinder D. R., British Journal of Political Science 11 pp 129– (1991) · doi:10.1017/S0007123400002544
[37] Kuhn D. P., Politico (2008)
[38] Lewis-Beck M. S., Forecasting Elections (1992)
[39] Lewis-Beck M. S., PS: Political Science & Politics 41 pp 687– (2008)
[40] Lock K., Political Analysis 18 pp 337– (2010) · doi:10.1093/pan/mpq002
[41] Lunn D. J., Statistics and Computing 10 pp 325– (2000) · doi:10.1023/A:1008929526011
[42] McDermott M. L., Public Opinion Quarterly 67 pp 244– (2003) · doi:10.1086/374574
[43] Nadeau R., The Journal of Politics 63 pp 159– (2001) · doi:10.1111/0022-3816.00063
[44] Panagopoulos C., Public Opinion Quarterly 73 pp 119– (2009) · doi:10.1093/poq/nfp010
[45] Panagopoulos C., Presidential Studies Quarterly 39 pp 896– (2009) · doi:10.1111/j.1741-5705.2009.03713.x
[46] Pew Research Center, Project for Excellence in Journalism (2008)
[47] Pickup M., International Journal of Forecasting 24 pp 272– (2008) · doi:10.1016/j.ijforecast.2008.02.007
[48] Pollster.com. 2008. ”The Polls: The 2008 Presidential Election”. Available athttp://www.pollster.com/polls/2008president[online]
[49] R Development Core Team, R: A Language and Environment for Statistical Computing (2011)
[50] DOI: 10.1080/10584609.1996.9963108 · doi:10.1080/10584609.1996.9963108
[51] Rigdon S. E., American Politics Research 37 pp 700– (2009) · doi:10.1177/1532673X08330670
[52] Romer D., Capturing Campaign Dynamics, 2000 and 2004: The National Annenberg Election Survey (2006)
[53] Rosenstiel T., Public Opinion Quarterly 69 pp 698– (2005) · doi:10.1093/poq/nfi062
[54] Rosenstone S. J., Forecasting Presidential Elections (1983)
[55] Shaw D. R., Journal of Politics 61 pp 387– (1999) · doi:10.2307/2647509
[56] Stevenson R. T., British Journal of Political Science 30 pp 217– (2000) · doi:10.1017/S0007123400000107
[57] Stovall J. G., Public Opinion Quarterly 48 pp 615– (1984) · doi:10.1086/268862
[58] Strauss, A. 2007. ”Florida or Ohio? Forecasting Presidential State Outcomes Using Reverse Random Walks,”. Working Paper: Princeton University.
[59] Sturtz S., Journal of Statistical Software 12 pp 1– (2005) · doi:10.18637/jss.v012.i03
[60] Traugott M. W., Public Opinion Quarterly 65 pp 389– (2001) · doi:10.1086/322850
[61] Traugott M. W., Public Opinion Quarterly 69 pp 642– (2005) · doi:10.1093/poq/nfi061
[62] Traugott M. W., The Voter’s Guide to Election Polls (2008)
[63] Traugott M. W., Public Opinion Quarterly 73 pp 866– (2009) · doi:10.1093/poq/nfp078
[64] Vavreck L., The Message Matters: The Economy and Presidential Campaigns (2009) · doi:10.1515/9781400830480
[65] Weisberg H. F., The Total Survey Error Approach: A Guide to the New Science of Survey Research (2005) · doi:10.7208/chicago/9780226891293.001.0001
[66] Wlezien C., American Politics Research 24 pp 492– (1996) · doi:10.1177/1532673X9602400406
[67] Wlezien C., American Politics Research 29 pp 419– (2001) · doi:10.1177/1532673X01029005001
[68] Wlezien C., The Journal of Politics 64 pp 969– (2002) · doi:10.1111/1468-2508.00159
[69] Wlezien C., International Journal of Public Opinion Research 19 pp 74– (2007) · doi:10.1093/ijpor/edl001
[70] Zaller J., The Nature and Origins of Mass Opinion (1992) · doi:10.1017/CBO9780511818691
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.